Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From the collection of the 



o Prelinger 

i a 

JJibrary 



San Francisco, California 
2007 




I 

JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 JANUARY, 1946 No. 1 

CONTENTS 

PAGE 

Citation on the Work of Charles J. Kunz, Herbert E. 
Goldberg, and Charles E. Ives G. E. MATTHEWS 1 

Intermodulation Distortion of Low Frequencies in Sound 
Film Recording F. G. ALBIN 4 

A System of Lens Stop Calibration by Transmission 

E. BERLANT 17 

Army Film Distribution and Exhibition 

R. A. KISSACK, JR. 26 

An Analysis of the Comparison of Beam Power and 
Triode Tubes Used in Power Amplifiers for Driving 
Loudspeakers J. K. HILLIARD 30 

Push-Pull Frequency Modulated Circuit and Its Appli- 
cation to Vibratory Systems A. BADMAIEFF 37 

A Discussion of the Acoustical Properties of Fiberglas 

W. M. REES AND R. B. TAYLOR 52 

A Three-Band Variable Equalizer L. D. GRIGNON 64 

Psychological and Technical Considerations Employed 
in the Bucky Sound Reproduction and Public Address 
Systems P. A. BUCKY 75 

Technical News 80 

Society Announcements 85 

Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and. December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCItTY of MOTION PICTURE ENGINEERS 

MOTCL PENNSYLVANIA NCW YOFSK !, N-Y- TCL. PSNN. 6 O62O 

HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 

^President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
* Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
*Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
** Engineer ing Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
^Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
**Financial Vice-President: M. R. BOYER, 

350 Fifth Ave., New York 1. 
* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
* Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
^Treasurer: EARL I. SPONABLE, 

460 West 54th St., New York 19. 

Governors 

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
**WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairman, Atlantic Coast Section. 
**Term expires December 31, 1947. ^Chairman, Pacific Coast Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 JANUARY, 1946 No. 1 



CITATION ON THE WORK OF CHARLES J. KUNZ, HERBERT 
E. GOLDBERG, AND CHARLES E. IVES* 



GLENN E. MATTHEWS** 



The Journal Award for 1945 has been made to Charles J. Kunz, 
Herbert E. Goldberg, and Charles E. Ives. Appropriate certificates 
were presented to them by President D. E. Hyndman on the evening 
of October 16, 1945, at the dinner-dance held during the Fifty- 
Eighth Semi-Annual Technical Conference of the Society. This 
award is given annually for the most outstanding paper originally 
published in the JOURNAL of the Society during the preceding calen- 
dar year. The title of the paper designated for this award is "Im- 
provement in Illumination Efficiency of Motion Picture Printers," 
published in the May 1944 issue. The paper describes the applica- 
tion of condenser and reflector systems for 2 typical printers with 
which increased illumination efficiency is obtainable. 

The three authors are members of the staff of the Kodak Research 
Laboratories. Charles E. Ives was born in Rochester, New York, 
and attended schools in that city. He has taken specialized training 
in the extension courses of the University of Rochester in the fields 
of chemistry, physics, mathematics, optics, and languages. Mr. 
Ives joined the Eastman Kodak Company in 1919 and worked from 
that date until 1940 mainly on the problems of motion picture 
laboratory operation. He carried out useful studies in several fields 
of motion picture engineering related to film processing, instrument 
design, printing, surface treatment of film and analysis of defects, 
tinting and toning, storage of film, and other related subjects. 

Mr. Ives' experience was also used to plan the organization and 
arrangements of the film developing department. In 1940, he became 
a supervisor in the photographic chemistry department and has 

* Recipients of 1945 Journal Award of the Society. 
** Technical Editor, Kodak Research Laboratories, Rochester, N. Y. 



2 , GLENN E. MATTHEWS Vol 46, No. 1 

continued work on specialized problems related to the progress of 
the war. During the 26 years of association with the Kodak Re- 
search Laboratories, he has published 25 papers in the Transactions 
and JOURNAL of this Society. He also served on several committees 
and assisted in the editing of the Transactions. 

Mr. Ives lives in an unusual 3-level modern home in Huntington 
Hills near Rochester and claims as his hobbies sailing, hiking, and 
listening to good music. 




FIG. 1. D. E. Hyndman (right), President of the Society, presents the 
1945 Journal Award to, left to right, Herbert E. Goldberg, Charles J. Kunz, and 
Charles E. Ives, co-authors of the paper, "Improvement in Illumination Effi- 
ciency of Motion Picture Printers." 

The second author, Herbert E. Goldberg, was born in Leipzig, 
Germany, and was educated in France. He received the degree 
Licencie es Sciences (B.Sc.) at the Sorbonne (University of Paris) 
in 1935. Subsequently, he continued his training in Paris and ob- 
tained degrees in optical and electrical engineering at the Institut 
d'Optique and the ficole Sup. d'electricite, respectively. Mr. Gold- 
berg joined the Eastman Kodak Company in November, 1937, and 
has worked since that date as a research engineer in electronics in 
the physics department. Throughout the war he carried on special 
work in this field. He collaborated in the preparation of an unusual 



Jan., 1946 CITATION ON WORK OF KUNZ, GOLDBERG, AND IVES 3 

article on "The Science of Color" published in Life for July 3, 1944. 
In private life, Mr. Goldberg is a great enthusiast for hiking and has 
spent many delightful hours on the Adirondack trails. 

The third of this trio of authors, Charles J. Kunz, was born in 
Rochester, New York. He graduated from Villanova College in 
1933 with the degree of Bachelor of Science in Chemical Engineering. 
After a year spent as a staff chemist with the American Chemical 
Products Company in Rochester, New York, he joined the film de- 
veloping department of the Eastman Kodak Company. His work 
in this department has consisted in the design of special developing 
equipment, effects of agitation in motion picture developers, and the 
standardization and handling of photographic materials during 
processing operations. The results of some of these investigations 
have been published in several papers in the JOURNAL. Mr. Kunz 
enjoys photography and gardening as hobbies. 



INTERMODULATION DISTORTION OF LOW FREQUENCIES 
IN SOUND FILM RECORDING 



FRED G. ALBIN* 



Summary. Variable-density sound film recording is subject to inter modulation 
distortion in the low-frequency band which extends from the lower limit of the audible 
spectrum toward zero, but not to zero frequency. This distortion is the result of the 
"Bromide Drag" phenomenon and is in the nature of exorbitant gamma. Cor- 
rections may be made for gamma at any point in this exorbitant range, but only at the 
sacrifice of correct gamma for other frequencies. 

The popular intermodulation method of distortion analysis uses 60 cps as a low- 
frequency component, which frequency lies in the exorbitant band. The data ob- 
tained thereby are representative of the exorbitant conditions and are misleading if 
they are taken to represent the entire recorded frequency spectrum. 

The delta-db method of distortion analysis gives data representative of conditions 
in which the low-frequency component is zero frequency. The analysis shows, 
however, that the data are also representative of frequencies above the exorbitant band 
and, therefore, represent the general recorded frequency spectrum except for the 
exorbitant band. 

Optimum processing specifications should be based upon data representing both 
conditions outside the exorbitant band, as indicated by the delta-db method (zero 
frequency], and inside the band, as indicated by the intermodulation method using 
several frequencies around 60 cps. 

Either method may be used alone for processing control data after all conditions 
have been determined and are known to remain constant. Practice has substantiated 
the merits of the principles expounded here. 

"BROMIDE DRAG" PHENOMENON 

Variable-density sound film processing, like processing of action 
photographic film, is subject to nonuniformity as the result of the 
release of the products of the reaction of the developer as it reduces 
the silver halide. The accumulation of these reaction products tends 
to retard the reaction by attenuating the concentration of the de- 
veloping reagents, and by a definite depressing effect on the reaction 
itself. This accumulation of reaction products streaks downward 
across a photographic plate in a still processing bath, and the effects 
are known as "Mackie" lines. In machine development, where the 

* Radio Corporation of America, RCA Victor Division, Indianapolis, Ind. 
4 



INTERMODULATION DISTORTION 5 

film is pulled longitudinally through the developing bath, the accumu- 
lation of reaction products is "dragged" to a trailing position from 
where it diffused from the emulsion. The accumulation of reaction 
products is considered as consisting effectively of bromide ions. 
Thus, the effect of this phenomenon is known as the "Bromide Drag 
Effect." 

The bromide drag effects, always present to a more or less serious 
degree, are influenced very extensively by the processing technique. 
The more important factors in the processing technique are the degree 
of developer solution agitation relative to the lineal velocity of the 
film through the developer solution, and the slope of the gamma- ver- 
sus-time curve at the point employed. In other words, minimum 
bromide drag effect is realized when the agitation of the developer 
solution is greatest, and when the gamma of development system 
nears its "infinity," or saturation; thereupon, the developing time 
produces a negligible increase of gamma. 

An easy demonstration 1 of the presence of bromide drag can be 
given by means of a sample of film prepared in the following manner : 
A mask is prepared to cover rectangularly shaped areas of the film. 
These rectangular areas have a dimension in the direction longi- 
tudinal with the film ranging from 1 / w in. upward, and are widely 
spaced in this direction. The unmasked area of the film is exposed 
with a medium value of exposure which, when developed to a gamma 
lower than its infinity, will have a density lying approximately at the 
point on the H and D curve where the slope is the greatest. A second 
and larger exposure is made by using a second mask which will allow 
exposures only of a series of rectangularly shaped areas similar in 
sizes and spacing to the first series. The second series is remotely 
spaced from the first series, but lies in the same field of medium ex- 
posure, and is heavily exposed. A third exposure about one- tenth 
as great as the medium exposure is given to the entire unmasked film. 

This sample of film is now developed by a developing machine, 
fixed, washed, and dried. The resulting sample consists of a field of 
medium density upon which are tablets of both small and large 
density. Inspection will reveal that the density surrounding these 
tablets is not uniform with the density in the field remotely located 
from these tablets. (See Fig. 1.) As the film passes through the 
developer solution, and the reduction of the silver salt takes place by 
the reducing action of the developer, by-products of the chemical 
reaction which consist effectively of bromide ions are released. The 



FRED G. ALBIN 



Vol 46, No. 1 



concentration of the bromide ions is in proportion to the amount of 
silver reduced, and, consequently, to the density of the image. In 
the field of uniform exposure the bromide concentration can be con- 
sidered as uniform, but surrounding the tablets of greater exposure 
there will be an increased concentration of bromide. This cloud of 
concentrated bromide will be dragged by the motion of the film 
through the developer, so that it trails behind the image of greater 
exposure. Consequently, in the area following this image will be an 



A+D 



r 



T 




BISECTION or Derfiope* MOMS ftifi. 



DlSTAMGf lONGITUDMAUr 

FIG. 1. Densities measured along film as a result of bromide drag in develop- 
ment. Exposures consist of tablets of each, high and low exposure, in a field 
of uniform medium exposure. 



area of reduced development where the concentrated bromide ions 
have retarded the development, the results being lower density. 

The same action as described results in the opposite effect in the 
area trailing the tablets of low exposure. In this instance, the area 
trailing the tablet of low exposure has bromide concentration lower 
than the average, which results in accelerated development in the 
region immediately following the tablet of low exposure. The tail 
may not necessarily be directly behind the tablet in line with the film 
travel. But it will be in a direction from the tablet determined by 
the resultant velocity of the developer relative to the film, determined 
by the film motion, and the prevailing motion of the developer 



Jan., 1946 INTERMODULATION DISTORTION 7 

owing to externally created agitation. The result of this bromide 
drag action is effectively to increase the contrast wherein there is a 
greater difference in density than normal between the tablet density 
and the -density of the field immediately adjacent thereto. As 
stated previously, this bromide drag is always present and the length 
of the tail following the tablet is inversely dependent upon the degree 
of agitation of the developer at the surface of the film. 

WAVE DISTORTION 

Now let us consider what occurs in the case in which we have, in- 
stead of isolated tablets, a repeated function in the form of a low- 
frequency wave. Such a wave on film consists of a succession of dark 
half -waves and light half -waves. As shown above, a dark half -wave 
is followed by a light tail and a light half -wave by a dark tail. If the 
length of a half -wave corresponds approximately to the length of the 
tail, a dark tail coincides with the following dark half-wave and a 
light tail coincides with the following light half- wave. Thus, the 
bromide drag has the effect of reinforcing the amplitude of the wave 
in a manner similar to increase in gamma. If the length of the half- 
wave is very great relative to the length of this tail, the tail coincides 
with only a small portion of the following half-wave, and the average 
increase in contrast for the sine wave diminishes accordingly. Like- 
wise, if the length of the wave is very short compared with the length 
of the tail, the tail coincides with both light and dark halves of the 
cycle, and no reinforcement occurs because there is no increase in 
contrast between the dark and light halves of the cycle. 

Experimental data to prove the foregoing theory were obtained by 
setting up apparatus wherein a variable-density recording machine 
was used in the conventional manner for sound recording so far as ex- 
posing intensity, mean exposure, type of film and processing condi- 
tions were concerned. The apparatus was set up in a manner similar 
to that described later in this paper as "Variable Frequency Inter- 
modulation Setup." Recordings were made of intermodulation test 
signals witfc the low-frequency component at discrete values between 
zero and 200 cps. 

Three such identical sets of recordings were made for processing at 
3 different film laboratories, A , B, and C: 

(A} Laboratory A used a processing machine with below-normal turbulence 
of the developer solution. 

(5) Laboratory B used a spray type of turbulence wherein the developer was 



8 



FRED G. ALBIN 



Vol 46, No. 1 



applied to the top ends of the loops of film as the film passed over the top rollers 
in the tank. The height of developer in the tank was considerably below the tops 
of the loops. The developer consequently had a motion whereby it ran down both 
sides of the loop from the top. Thus, as the film passed through the machine, the 
direction of the developer was first one way, then the opposite, along the film. 

(C) Laboratory C used relatively vigorous agitation together with a de- 
veloper formula which required a relatively long time of development, thus 
operating at a point of low-slope on the time-gamma curve. In other previous 
recordings, the speech and music quality of this processing was judged the best 
available. 

(Z>) Typical commercial laboratory on a par with A. 

After processing and printing, the intermodulation recordings were 
reproduced through a review room reproducing system which in- 




20 40 60 80 
FREQUENCY CAS. 



FIG. 2. Intermodulation versus frequency of low-frequency 
component: A from film processed by laboratory A; B from film 
processed by laboratory B; C from film processed by laboratory C. 



eluded a 1000-cps high-pass filter especially for this occasion. The 
levels were read by means of a special volume indicator consisting of a 
high-speed galvanometer wherein the displacement was linear with 
the instantaneous amplitude of the distorted wave. The variations 
of 1000-cps amplitude were taken as the data for computing distor- 
tion. By this arrangement, true peak values of distortion irrespec- 
tive of frequency and wave-form were obtained, using Fo^nula 1. 

Intermodulation apparatus in current use usually employs an 
indicating meter of the copper-oxide rectifier type which gives a read- 
ing approximately proportional to the average value of the distortion 
wave. The distortion wave is seldom sinusoidal. Certain equip- 
ments in current use employ a vacuum tube voltmeter circuit for the 
level meter which circuit has a response more nearly proportional to 



Jan., 1946 INTERMODULATION DISTORTION 9 

the peak value of the distortion wave. Under identical conditions of 
distortion, the type employing the copper-oxide rectifier type meter 
gives the lowest indication of distortion, the vacuum tube voltmeter 
type gives the next greater, and the delta-db or the high-speed oscil- 
lograph indicator type both indicating the maximum or the peak 
value of distortion. The differences between these indicator values 
for the same sample of distortion are very large, which account for 
the fact that Fig. 2 displays a case of 50 per cent distortion, which 
would read in the order of 12 per cent by the more familiar copper- 
oxide rectifier type indicator. Fortunately, as the distortion is 
lessened, the readings of the 3 types of meters converge, all reaching 
zero distortion simultaneously. Furthermore, since it is more im- 
portant to determine the conditions for minimum distortion, the 
absolute distortion value is unimportant except for record purposes. 

The conclusions to be derived from the data obtained as illustrated 
by Fig. 2 are that the measured intermodulation distortion is a func- 
tion of the frequency of the low-frequency component present. 

An early method of determining optimum film processing is popu- 
larly known as the Delta-db Test. 2 This test is a special intermodula- 
tion test wherein the low-frequency component is zero frequency. It 
consists of the recording of a 1000-cps tone of low amplitude two or 
more times, using a different value of d-c bias each time. This test 
has enjoyed popular favor for the following reasons : 

(1) It was the first over-all dynamic check of film processing in use. 

(2) The test can be applied without any special apparatus other than that 
which is part of the regular recording system, namely, a test oscillator to produce 
the high-frequency signal, and a bias current supply with a key for on and off, or 
polarity reversing. 

(5) The test can be reproduced through a standard theater or review room 
reproducing system with no special apparatus other than an ordinary volume 
indicator such as is used for usual level measurements. 

(4) The test indicates the peak value of distortion for the particular exposure 
excursion that corresponds to the amplitude of the low-frequency component. 
In this case, double amplitude of the low-frequency component is represented by 
the difference between the bias exposure and the unbiased exposure, or between 
any 2 bias exposure values which may be chosen. 

(5) A series of readings of peak distortion versus amplitude may be obtained 
by simply choosing a series of bias current values at which the exposures are made: 

(a) From any 2 readings taken with 2 values of bias, the effective over-all 
gamma, over the latitude defined by these 2 negative exposure values re- 
sulting from the 2 values of bias, may be computed very readily. (See 
Formula 2.) 

(6) The toe or shoulder characteristic of either negative or positive may be de- 



10 FRED G. ALBIN Vol 46, No. 1 

termined from 2 delta-db readings, one of which lies on the mid range of 
exposure, and the other of which lies on either the toe or shoulder. The 
effective over-all gamma for this region, when calculated, is a measure 
of the slope of the toe or shoulder. Corrections as desired may be made 
by shifting the mean density, thereby shifting the density range employed 
for recording either on or off the toe or shoulder as conditions require. 

Investigation as to the causes of film distortion discloses the 
following possible contributors : 

(1 ) Recording amplifier and system nonlinear characteristics. 

(2) Recording modulator device nonlinear characteristics. 
(5) Film latent image spread of negative. 

(4) Negative gamma : 
(a) Average value. 

(b} The instantaneous variation of nonlinearity of H and D curve. 

(5) Bromide drag or Eberhardt effect on negative. 

(6) Printer distortion. 

(7) Latent image spread of positive. 

(8) Positive gamma : 

(a) Average value. 

(b) Instantaneous variation or nonlinearity of H and D curve. 

(9) Bromide drag on positive. 

(10) Reproducer contrast coefficient. 
(11} Photoelectric cell nonlinearity. 

(12} Reproducing amplifier and system-nonlinearrty characteristics. 

* 

Items 1, 2, 4, 8, 10, 11, and 12 may be considered as having non- 
linear characteristics independent of the frequency of the low-fre- 
quency component of the intermodulation test. The remainder can 
be proved to introduce distortion with magnitude dependent upon the 
frequency. 

The 2 items which are probably the contributors of the greatest 
distortion dependent upon frequency are items 5 and 9, and the 
figures given show that the magnitude of this distortion is consider- 
able. The remainder of the discussion will be confined largely to 
these 2 items. It might be mentioned that the existence of these 12 
sources of possible distortion justifies the intermodulation method of 
distortion measurement and makes inadequate any static method of 
checking of film processing. 

Another interesting sidelight is that any one of the 12 contributing 
sources measured alone might indicate a distortion higher than the 
aggregate of the twelve. This would indicate that there is a very 
decisive compensating effect present. Some of these sources are very 
obvious and well known to film technique. For example, it is com- 



Jan., 1946 INTERMODULATION DISTORTION 11 

mon knowledge that small variations of negative gamma (item 4a) 
can be compensated for completely by proper adjustment of positive 
gamma (item 8d). It also happens that items 10 and 11 have effects 
very similar to items 4a and 8a, so that the nonlinearity of the 
aggregate of these four may readily be canceled out. 

RECORDING PRACTICE 

In recording practice for motion pictures, the frequency spectrum 
lies between 50 cps and 8000 cps, these limits being established by 
appropriate high-pass and low-pass niters. The attenuation by the 
high-pass filter for frequencies below 50 cps is very large, hence the 
signal energy below this cutoff point is negligible. However, the 
action of ground noise-reduction facilities is to impress upon the film 
exposing illumination a modulating frequency equivalent to the 
envelope frequency, which frequency extends down to zero. In view 
of this condition, the effect of the low-frequency component lying in 
the band between zero and 60 cps might very well be considered. 

As all frequencies ranging from zero cps upward are encountered in 
recording practice, the question arises as to what frequency should 
be used for testing intermodulation distortion. 

The intermodulation test which is in popular use employs a single 
low frequency of 60 cps for the low-frequency component, which 
frequency lies definitely within the range of exorbitant gamma. The 
results obtained thereby are representative of conditions at the one 
frequency only and not of the average conditions for the recording 
spectrum. It would seem probable that the preferred frequency 
would not lie in the range from 30 to 60 cps, which frequencies lie 
near the region of maximum distortion for a typical laboratory, 
(A or D). Neither would it seem logical to choose a frequency 
above 100 cps. It would, of course, be very desirable to improve 
the film processing technique so that the distortion, as illustrated by 
Fig. 2, could all be held to very low values; in such a case the choice 
of frequency would be unimportant. 

It is apparent that the delta-db test is also inadequate for providing 
the information as to conditions within the frequency band where the 
bromide drag effect is prevalent and the gamma is exorbitant. In 
the present state of the art, the curves for laboratories A and D 
represent typical conditions and although not desirable, do prevail, 
and a choice of test frequency must not be made arbitrarily. There is 
no choice but to explore the low-frequency region to determine the 









12 FRED G. ALBIN Vol 46, No. 1 

conditions prevailing, and strive to obtain a compromise for optimum 
over-all processing conditions. 

VARIABLE FREQUENCY INTERMODULATION SETUP 

Signals of frequencies much lower than 50 cps cannot be passed 
satisfactorily by the speech transmission circuits. The ground noise- 
reduction circuits will pass signals of frequency extending down to 
zero cycles, and may be used as a means of communicating low- 
frequency modulations to the film. An amplitude-modulated 1000- 
cps signal may be generated at the input of the ground noise-re- 
duction amplifier by beating together 2 signals of frequencies that 
differ by the amount of the low frequency required. The ground - 
noise rectifier system acts as a detector to produce the required low 
frequency, which is then applied to the shutter, valve, or other device, 
normally used as a mean exposure control for noise-reduction purposes. 

A suggested equipment setup is outlined as follows : 

(1) Set up the recording system in the normal manner with noise-reduction 
facilities. Set the exposing lamp intensity so that the exposure range with full 
modulator excursion lies within the linear latitude of the film. 

(a) Full modulator excursion will be referred to as 100 per cent modulation, 3 
with peaks extending to zero and 100 per cent exposure. 

(b) The mean (unbiased) exposure is 50 per cent. 

(c) "Bias" mean exposure, for noise reduction, is referred to by its ratio to the 
unbiased exposure. Noise-reduction decibels are 20 times the common 
logarithm of the ratio of unbiased-to-bias exposures. 

(2} Adjust the system to allow a minimum of 6-db "reverse bias." 4 This 
means that as the input level to the ground noise-reduction amplifier is increased, 
the mean exposure will increase linearly until at least 100 per cent is reached. 

(3) Set the bias for 4.7-db noise reduction. 

(4) Set the gain of the ground noise-reduction amplifier for 12-db margin. 

(5) Introduce a signal of 1000 cps into the system with a level of 16 db rela- 
tive to 100 per cent modulation. 

(6) Introduce a second signal into the input of the ground noise-reduction 
amplifier (but not into the recording amplifier) in a manner which adds the 
amplitude of the second signal to the first. The amplitudes of the first and 
second signals at the point where they are added must be equal if the specified bias 
and levels are used. The wave-form of the low-beat frequency would be more 
sinusoidal if these 2 signals were widely different, requiring new bias and level 
values. The frequency of the second signal shall be variable from the same fre- 
quency as the first (1000 cps) to 200 cps greater. 

(7) The wave filter normally incorporated in the ground noise-control appara- 
tus would effectively limit the low-frequency component to a frequency below 
30 cps. Therefore, it is necessary to replace this filter with one with a higher 
cutoff frequency. It is important to retain a filter with cutoff frequency below 



Jan., 1946 INTERMODULATION DISTORTION 13 

1000 cps so as to eliminate the 1000-cps signal. The cutoff frequency of this filter 
should preferably be not greater than an octave above the low frequency so as to 
remove the sidebands and leave only the fundamental sine wave of low frequency. 
Thus, the wave produced by beating the 2 signals need not necessarily be 
sinusoidol. 

The above described setup will apply to the film an intermodulation 
test consisting of a 1000-cps signal superimposed on a low-frequency 
signal of 12-db higher amplitude. The peak excursions of the com- 
plex wave lie between the limits of 10 per cent and 90 per cent ex- 
posure, which is as great an excursion as should be used for inter- 
modulation tests. The low-frequency signal may be varied in fre- 
quency at will between zero and 200 cps, for initial tests of new 
processing conditions. Afterward, the range may be narrowed 
down, eventually resolving to a single frequency, such as zero or 60 
cps which can be obtained more readily. If the relationship be- 
tween characteristics may be safely assumed to remain fixed, any 
point may be chosen as a control point for the routine checking of 
processing. Recordings are made at discrete low-frequency values 
from zero to 200 cps. 

The resulting recordings are processed and printed in the normal 
manner and reproduced in a standard sound reproducing system. 
The output levels of the recordings made with extremely low fre- 
quencies may be read on a standard high-speed rectifier-type volume 
indicator, no special analyzer, filter, etc., being required. The re- 
producer amplifier system will not pass frequencies much below 30 
cps and it therefore acts as though a high-pass filter were included. 
Conditions free from distortion will give a steady reading on the 
meter, since the meter reads the unmodulated 1000-cps tone. The 
amplitude of the excursion of the meter is a measure of the inter- 
modulation distortion. 

Formulas 1 and 2 may be used to compute the harmonic distortion 
and effective over-all gamma, respectively, where R\ and Rz are the 
minimum and maximum meter readings, in decibels, of the fluctuating 
meter readings. 

Where RI = VI reading (db) at exposure E 
R 2 = VI reading (db) at exposure E 2 

D D 

Per Cent Harmonic Distortion (peak value) = antilog X 100 per cent (1) 

n p 

Effective Over-all Gamma = 1 l (2} 

20 Iog 10 - 2 

tL\ 



14 FRED G. ALBIN Vol 46, No. 1 

If the suggested setup values were used : 

Eo 90 

= = 9 and 20 log 9 = 19.1 db. 

-C*i 1U 

In determining effective over- all gamma, 5 it is important to 
identify R and ^2 with respect to E\ and E z . Here the advantage of 
the use of zero frequency is obvious, wherein Ei and E 2 are recorded 
in a prechosen sequence and RI and R^ are read in the same sequence. 

When the low frequency is near to, or higher than, the resonance 
frequency of the indicating meter, the readings are not reliable. For 
these higher frequencies, a cathode-ray oscilloscope together with a 
500-cps high-pass filter may be used as an indicator to replace the 
rectifier-type meter. The resulting wave will be a 1000-cps wave 
modulated in amplitude in accordance with the magnitude of the dis- 
tortion. The amplitudes may be scaled from the cathode-ray oscil- 
loscope image, and the distortion computed therefrom. 

CHOICE OF CONTROL FREQUENCY 

That component of intermodulation distortion which results from 
improper contrast or gamma can be canceled completely; for in- 
stance, by proper choice of negative gamma. Thus, processing con- 
ditions may be chosen to result in minimum distortion at any arbi- 
trarily chosen low frequency. It would seem logical to choose condi- 
tions which will reduce the average distortion to minimum, under 
which the greatest departure from the minimum would be reduced. 

By extensive listening test, wherein the processing was regulated to 
give minimum distortion at the points lying between the distortion 
zero cps and 60 cps, it has been determined that the optimum condi- 
tion has been obtained lying approximately midway between zero and 
60 cps. Such a frequency is odd, and cannot be obtained as readily 
as zero or 60 cps are now obtained with existent facilities. Instead 
of choosing such a frequency "for routine checking at processing, it is 
perfectly practical to make the measurement at either zero or 60 cps, 
and interpolate the data for the case represented by this intermediate 
frequency. This practice is now being followed by certain Hollywood 
studios. As pointed out earlier in this discussion, the distortion evi- 
dent at 60 cps as compared with zero cps, is equivalent to the distor- 
tion resulting from excessive contrast or gamma. With this condi- 
tion known, it is perfectly practical to make the measurement at zero 
cps, but with an over- all gamma slightly lower than optimum for the 



Jan., 1946 



INTERMODULATION DISTORTION 



15 



frequency. This correction is in the order of 10 per cent. Also, it is 
perfectly practical to make the measurement at 60 cps and use an 
over-all gamma slightly higher (10 per cent) than the optimum for 
this frequency. This technique has been followed in practice with 
good results. (See Figure 3.) 



FkfQUFNCY C./?S. 



Intermodulation versus frequency of low-frequency component 



Film from laboratory B; process adjusted for lowest average intermodula 




tion distortion 



. REFERENCES 

1 LESHING, M., INGMAN, T., AND PIER, K.: "Reduction of Development 
Sprocket Hole Modulation," /. Soc. Mot. Pict. Eng., XXXVI, 5 (May, 1941), p. 
475. 

2 ALBIN, F. G.: "A Dynamic Check on the Processing of Film for Sound 
Records," J. Soc. Mot. Pict. Eng., XXV, 2 (Aug., 1935), p. 161. 

3 The reference to 80, 100, or any percentage modulation of the light modulator 
has little significance. Reference to per cent modulation of variable-density 
film has no significance at all. If a modulator such as a light valve, deflected 
penumbra, or glow lamp, has fixed average intensity, the modulation may be 
then stated definitely in terms of percentages of complete or 100 per cent modula- 
tion. In this case, 100 per cent modulation is that level at which the amplitude 
of the wave is equal to the average or mean value of light intensity in which case 
the excursion of light intensity extends from zero to double the mean. 

The use of bias (for ground noise reduction) and reverse bias (for level range 
extension) has complicated this concept. However, for a reference point, some 
arbitrarily chosen mean value of light intensity is established, and the percentage 
modulation is referred to this. The use of this mean value and 100 per cent 
modulation thereof imposes upon the film an excursion of exposure which the film 
may accommodate with reasonably low distortion. This is the customary proce- 
dure in studio recording practice. The percentage modulation values referred to 
herein are so reckoned. 



16 FRED G. ALBIN Vol 46, No. 1 

4 HANSEN, E. H., AND FAULKNER, C. W. : "Mechanical Reversed-Bias Light- 
Valve Recording," /. Soc. Mot. Pict. Eng., XXVI, 2 (Feb., 1936), p. 117. 

5 The over-all effective gamma determined in this manner is the gamma of a 
simple curve, 



having only a single value of gamma, which curve includes the 2 points used for 
the computation of gamma. Generally, film characteristics as used in practice 
involve not only the straight portion of the H and D curve, but the toe and 
shoulder as well. The positive film H and D toe region is extensively used in 
variable-density sound recording. The curve involving both the straight line 
and toe portions requires multiple values of gamma for expression. To express 
such a curve in terms of gamma, and to take delta-db data to compute gamma, 
requires that the curve be scanned in short increments. The gamma for each 
increment as defined by 2 exposure values should be calculated by use of Formula 
2. 

In practice it has been learned that the over-all effective gamma should not 
exceed unity to any appreciable extent in any portion of the film characteristic 
curve. Gamma in excess of unity is equivalent to expansion, and is particularly 
deterimental to speech intelligibility. Compression in the same degree is accept- 
able. Thus, if low and medium level recordings, made with ground noise 
reduction, are confined to the straight portion of the H and D curve (which 
portion has an over-all effective gamma of unity or only slightly higher) , the higher 
level recordings will extend into the region of the toe of the positive H and D 
curve wheie the gamma is lower and some compression is encountered. This is 
the preferred condition of variable-density recording technique. 

For routine checking, 3 delta-db control points are required ; the first at bias 
value of exposure, the second at unbias, and the third at reverse bias. The 
gamma as computed from the first two should lie between unity and 5 per cent 
greater. The gamma as computed from the second and third should lie between 
unity and 10 per cent less than unity. 

The ability to check these 2 ranges separately is a unique advantage enjoyed by 
the delta-db test over the intermodulation test. For this paper, the term "Effec- 
tive Over-all Gamma" has been adopted to represent the exponent "of the over-all 
response curve, which is plotted with the current in the modulator as the abscissa 
and the photocell output current as ordinates. Effective over-all gamma thus 
includes all factors in the system which affect contrast or linearity, and is the 
figure which must be reduced to unity for linear performance of the system. The 
use of the term gamma in this manner has aroused some criticism, but the criti- 
cism is based upon academic rather than practical considerations. 



A SYSTEM OF LENS STOP CALIBRATION BY 
TRANSMISSION* 



EMMANUEL BERLANT** 



Summary. An apparatus is described for the measurement of relative trans- 
mission of photographic objectives, consisting of a controlled light source, condensing 
system, lens mount, integrating sphere, photo cell, and meter. Transmission char- 
acteristics of 100 typical objectives are presented, together with a proposed standard 
method of calibration that will reconcile present practice and equipment with future 
problems involving low reflectance coatings and objectives designed with high index 
glasses. 

With the increasing accuracy of exposure required by color photog- 
raphy and closely controlled monochromatic photography, interest 
in illumination characteristics of lenses has increased to the point 
where need for a system that would eliminate transmission varia- 
tion from lens to lens has become manifest. Any system proposed 
for general adoption must be based upon : 

(a) A standard of calibration reproducible from a mathematical premise, 
(6) Correlation to current calibration practice, to avoid disturbance of film 

speed data and enable use of the new system with exposure meters and exposure 

data now in existence, and 

(c~) Correlation to geometrical aperture ratio calibration, to permit use of 

depth of field tables, and other pertinent data based upon geometrical aperture 

ratios. 

It is well known 1 that the illumination of an image is indicated by 
the formula 

E = kB sin 2 e 

wherein B = brightness of object plane 

6 = angle formed by edge ray and axial ray of lens 
k = transmission loss factor 

The geometrical ratio of focal length to entrance pupil now used 
as a method of calibration fails to take into account variation of the 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
** Research Laboratory, Signal Corps Photographic Center, Long Island City, 
New York. 

17 



18 



EMMANUEL BERLANT 



Vol 46, No. 1 



loss factor k. This loss factor not only varies from design to design, 
but from lens to lens of the same design and manufacture. Further 
consideration must be given to the angle of coverage involving lenses 
of different focal length. For detailed consideration of the impor- 
tance and extent of illumination variance resulting from this cause 
alone, reference is made to G. Slussareff 2 and F. Benford. 3 

A photographic objective of large aperture also shows a marked 
deviation from the theoretical relationship existing between various 
geometrical apertures, owing to the change of surface reflection losses 
with the change of angular presentation of various zones of the lens. 
The extent of this deviation will be indicated in another part of this 
paper. 




FIG. 1. View of apparatus. 



Photometric methods of calibration may be used to integrate 
these phenomena and permit the matching of transmission values of 
various lenses of various focal length, with provision for the best 
balance of axial zone to edge zone levels of illumination commensurate 
with good practice. 

A method of photometric calibration described by D. B. Clark and 
G. Laube 4 involved the selection of a lens which is arbitrarily de- 
nominated as having a certain effective stop value wide open, based 
upon its geometrical ratio of focal length to effective aperture. Light 
from a diffuse plane surface is passed through the lens, and limited 
by an academy aperture in the focal plane of the lens. The lens is 
then calibrated by decreasing the illumination by an amount in ac- 



Jan., 1946 LENS STOP CALIBRATION BY TRANSMISSION 19 

cordance with the theoretical mathematical progression of //stop 
transmission by closing the diaphragm, and marking the indicated 
position of the diaphragm ring. 

This method results in a consistant progression of integrated il- 
lumination value, and other lenses may be matched to this standard. 
However, it is obvious this method suffers from the necessity of refer- 
ence to the arbitrary standard selected, and is therefore not feasible 
of extension beyond reach of the standard lens, except by the use of 
secondary standards, always an objectionable technique. A method 
described by E. W. Silvertooth 5 is based upon measurement of 
axial intensity, and uses as a basis of calibration an arbitrary aper- 
ture selected to correspond to the focal length of the lens being cali- 
brated, and making an arbitrary allowance for losses to be expected 
within the system. While this method permits reconstruction of the 
standard selected, it requires creation of a new standard for each 
focal length, and makes no provision for extra axial illumination 
characteristics. 

It was decided that any standard proffered by this laboratory 
would have to be based upon a method of absolute transmission, and 
the problem was attacked from that direction. 

Extensive use was made of work on transmission losses of optical 
systems as described by D. B. McRae, and the apparatus used was 
developed as an extension of that used in his investigation of trans- 
mission losses (Fig. 1). 

A 1000-w projection lamp was coupled to a voltage controlled line, 
and a beam of light condensed upon an opal glass masked by an 
academy aperture. The light from this aperture was admitted into 
an integrating sphere, and measured by a vacuum photocell and 
precision microammeter (Fig. 2). 

The lens to be measured is placed into a turret-type mount bored 
out to receive several different types of lens mounts, with a proper 
registration depth to the plane of the aperture (Fig. 3). The light 
passed through the lens is then admitted into the integrating sphere, 
where it is measured as a proportion of the total flux at the aperture. 

One hundred lenses of the highest quality, of all focal lengths from 
35-mm to 165-mm, were measured in this apparatus, and the results 
plotted as given in Fig. 4. The ordinate is the //stop as calibrated by 
the manufacturer, and abscissa is the transmission of the lens in terms 
of thousandths of unity of aperture flux, as measured by the described 
apparatus. 



20 



EMMANUEL BERLANT 



Vol 46, No. 1 




Jan., 1946 LENS STOP CALIBRATION BY TRANSMISSION 



21 



Of immediate note is the wide spread of transmission values for 
various lenses of the same calibrated //stop. This is not because of. a 
difference in focal length alone, as may be verified by reference to 
Table 1, wherein lenses of short focal length are seen to have equal 
and higher transmission values than lenses of longer focal length. 
Some of these lenses showing extreme values for a given calibration 
were checked by the Steinheil method 6 to verify their geometrical 
ratio of focal length to aperture, and were found to be of good com- 
mercial accuracy. 

The average transmission value 
for each group of //stops is indi- 
cated in Fig. 4 as "Mathematical 
Average/' It is of interest to 
note that a \/2 progression from 
the smaller //stops toward the 
larger //stops indicates a failure 
of the geometrical aperture ratio 
calibration method, owing to the 
progressively greater surface re- 
flection losses for outer zones of 
lenses of larger relative aperture 
and accords with observations on 
the subject made by L. C. Mar- 
tin. 7 

A. C. Hardy and F. H. Perrin, 8 
in their discussion of the photom- 
etry of optical systems, note that 
a photographic objective should transmit about one per cent of the 
object brightness at //8. This point has been indicated in Fig. 4. 
Measurements with this apparatus indicate an average transmission 
of one per cent for //4.7. A line based upon a slope angle of \/2 
progression drawn through this point is offered as a basis for pho- 
tometric transmission calibration. This standard is in accordance 
with actual use, and will not require revision of film speeds now in 
use, nor will it obsolete exposure meters now in use. Tables of ex- 
posure based upon published data are also valid for use with this 
system. 

It is advocated that lenses be marked with the geometrical //stop 
value in its usual place upon the manufacturer's identification ring, 
and the diaphragm setting ring be calibrated in accordance with the 




FIG. 3. Turret-type lens mount. 



22 



EMMANUEL BERLANT 



Vol 46, No. 1 



BSS 






i 



//r 

7 - 







D 
CE POI 
% TRA 



Is 



o 

2 



Jan., 1946 LENS STOP CALIBRATION BY TRANSMISSION 23 

//stop, or transmission value, shown in Fig. 5. This value is based 
upon a -\/2 progression through the point one per cent of transmission 
equals //4.7. This indication of the relationship existing between 
transmission value, or //stop, and focal length to effective aperture 
ratio, or //stop, will function as a basis for a correction factor in using 
tables of depth of field, etc., where greater than normal accuracy is 
required. 



3.5 
7.6 
4.8 
5.6 
4.2 
3.2 
2.5 
2.9 
3.7 
3.0 
4.5 
4.6 
3.8 
2.9 
4.8 
2.4 
4.1 
5.4 
3.1 
3.9 







TABLE 1 




Transmission in 


Thousandths of Unity for 


Calibrated f/ Stops 


Lens No.* 


Mfr. 


Focal Length in Mm 


f/2.8 


1 


A 


100.0 


24.8 


2 


B 


75.0 


45.7 


3 


B 


75.0 


32.4 


4 


B 


75.0 


31.9 


5 


C 


75.0 


26.7 


6 


C 


75.0 


25.7 


7 


C 


75.0 


20.9 


8 


A 


75.0 


25.2 


9 


A 


75.0 


26.7 


10 


C 


52.0 


26.7 


11 


C 


52.0 


27.6 


12 


B 


50.0 


31.9 


13 


B 


50.0 


26.7 


14 


C 


47.0 


19.0 


15 


D 


41.0 


20.0 


16 


A 


40.0 


19.5 


17 


B 


37.5 


23.8 


18 


A 


35.0 


29.0 


19 


A 


35.0 


22.8 


20 


A 


35.0 


21.7 



* These 20 lenses are selected at random from 100 lenses measured. 



The validity of this approach is verified by the measurement of 
lenses calibrated by Technicolor to an arbitrary standard, which 
paralleled this standard within slightly more than Y 4 -//stop, and 
measurement of a lens calibrated by Twentieth Century-Fox to an 
arbitrary standard, which fell wholly within the V4-stop tolerance 
advocated for accuracy of calibration. 

In summation, it is believed that adoption of this //stop system 
will permit universal calibration of lenses, coated or uncoated, in a 
manner that will enable their use with current emulsion speed values 
and exposure meter equipment, with a minimum of mental adjust- 



24 



EMMANUEL BERLANT 



Vol 46, No. 1 

































00 01 






U 
























/ 


-00 6 






























if. 


001 




u 


























/// 


-009 

S 




s 








^; 


















I // 






z 









ib 


















/// 








Cfl 
cr 








(O 

^ i 
















t 


1 






cr 

z 






T'l 


















'// 








ct 




a: 


1 


i S 
















/// 








1- 




cc 


*. 


















! // 










cr 


^ 




PJ 
















I // 














z ** 


3 I 
















I / 








( r 





z 


















x 


if 










i 


1 


y i 


5 S 














/ 


// 








u. 






5 H 
















//^ 










c 

1- 






< 31 


sl 














// 










C 
a 


i 




* 

S 


ftt 












L 


^ 


"z 


















z o ^ 








/ 


'// 


^- 
z 


2 
to 

CO 

2 




















!|! 




T 1 


M. 




a 

UJ 


CO 

z 




.01 














- 1 ^ 






/, 




^~ 


o 

z 


E 




.8 














p" 




2 


// 




\ 


s 


"C 




















] 


^ 








Ul 


t 






















^ 








\tt 


1 




















V 










' 




















I .( 


' 






























II 
































, / 1 


















-6'0 














I /i 
































f // 






























/ 


/I 






























/, 


'l 






























/ // 
































22 




















-C*0 










//; 


/. 




























.^ 


f 




























/ 


























6 0*0 






; t 


/ 






























ft 






























; 


>j 






























/ 































I 



o ^ 



s s s s s 



H 38 WON 



Jan., 1946 LENS STOP CALIBRATION BY TRANSMISSION 25 

ment, and a maximum of exposure accuracy. It eliminates confu- 
sion relative to allowance for exposure in the case of coated objectives, 
yet retains a basis for computing accurately optical relationships 
based upon //stop values. 

In the case of objectives intended for use other than with motion 
picture apertures, it is suggested the plate size to be covered be used 
as a limiting aperture, in order to establish an optimum exposure 
value for all parts of the negative. 

I wish to express appreciation to the personnel of the Pictorial En- 
gineering and Research Laboratory Division of the Signal Corps 
Photographic Center, New York, for their participation in the prepa- 
ration of this paper. 

REFERENCES 

1 McRAE, D. B. : "The Measurement of Transmission and Contrast in Optical 
Systems," /. Opt. Soc. Am., 33, 4 (Apr., 1943), p. 229. 

2 SLUSSAREFF, G.: "Brightness of the Image Formed by Wide-Angle Objec- 
tives," J. Phys. U.S.S.R., (1941), p. 537. 

3 BENFORD, F.: "Illumination in the Focal Plane," /. Opt. Soc. Am., (May, 
1941), p. 362. 

4 CLARK, D. B., AND LAUBE, G. : "Twentieth Century Camera and Ac- 
cessories," /. Soc. Mot. Pict. Eng., XXXVI, 1 (Jan., 1941), p. 50. 

6 SILVERTOOTH, E. W. : "Stop Calibration of Photographic Objectives," /. Soc. 
Mot. Pict. Eng., XXXIV, 2 (Aug., 1942), p. 119. 

6 LOCKETT, A.: "Camera Lenses," Isaac Pitman & Sons, Ltd. (London), 
1938, p. 46. 

7 MARTIN, L. C.: "An Introduction to Applied Optics," Isaac Pitman & 
Sons, Ltd. (London), 1932, p. 206. 

8 HARDY, A. C., AND PERRIN, F. H.: "Principles of Optics," McGraw-Hill 
Book Co. (New York), 1932, ch. 19. 



ARMY FILM DISTRIBUTION AND EXHIBITION* 
ROBERT A. KISSACK, JR.** 

Summary. The U. S. Army has become the largest distributor of motion pic- 
tures in the world. This paper describes the system of exchanges which handled 
distribution and reviews the contributions made by motion pictures to the war 
effort. 

The success of the Army film program is in large measure attributable to the co- 
operation given by the motion picture industry and the motion picture engineers. 

The Army has become the largest distributor of motion pictures in 
the world. This should not be a challenging statement to make even 
here in the capital of the motion picture industry, because it has been 
with the cooperation of the motion picture industry that Army film 
distribution and exhibition is so world wide in scope. 

Army film work is the responsibility of the Army Pictorial Service 
whose primary mission is the production and distribution of motion 
pictures, film strips, and still pictures. 

Army Pictorial Service motion picture production includes all 
training films authorized by the Army Service Forces and the Army 
Ground Forces and all orientation, information, and education films 
required by the Army. Industrial incentive, historical, and cam- 
paign films are produced at the direction of the Bureau of Public 
Relations, while certain special and foreign language films are pro- 
duced for the General Staff. 

Since the Army Pictorial Service produces films at the direction 
of other War Department agencies, no completed picture may be 
released without the approval of the initiating agency. The re- 
quirements of such agency also determine the method and extent of 
distribution of completed motion pictures, while the number of prints 
distributed for showings within the Service Commands and Theaters 
of Operation is based on the requirements of the respective Com- 
manding Generals. 

Many completed pictures are intended for release outside the 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
** Major, SC, Distribution Division, Signal Corps Photographic Center. 
Long Island City, N. Y. 
26 



ARMY FILM DISTRIBUTION AND EXHIBITION 27 

Army. Some are shown to workers in war plants throughout the 
country and are released with the approval of the Bureau of Public 
Relations. Other pictures intended for general public information 
and civilian morale are released through the War Activities Com- 
mittee of the motion picture industry. 

When the war started the accent was on training, and the job of 
film distribution was to cooperate in that huge task of starting to 
train an army by the quick and efficient visual aids methods of 
training films and film strips. First a system of film exchanges, or 
film libraries, was organized throughout the Service Commands of 
this country, a system which included 10 central exchanges and over 
200 subexchanges. Personnel were trained and assigned to these 
libraries as visual aids coordinators, and procedures for film dis- 
tribution, projection, and film exchange operation were standardized 
and incorporated in Army supply procedures. Military instructors 
were assisted by traveling teams of visual aids experts, and by catalogs 
and publicity material on new films. Research studies were instituted 
which provided highly important operating statistics on projection 
equipment and print utilization. 

As the Army went overseas, similar training film exchanges were 
set up in all overseas theaters. Meanwhile, the Overseas Motion 
Picture Service was handling the distribution of 16-mm prints of 
entertainment films contributed by the motion picture industry for 
free showing to U. S. troops overseas. Today, the film libraries and the 
film and equipment exchanges form a coordinated distribution net- 
work over the entire globe serving the Armed Forces with a wide 
variety of films. 

For example, each week a special version of the latest combat 
footage received from all fighting fronts is rushed to the General Staff, 
to the Commanding Generals of every overseas theater, and to the 
Commanding Generals of the Service Commands. 

Each week prints of a film called Combat Bulletin are circulated 
overseas and domestically as a military newsreel, a factual document 
of great interest to officers and troops of all services. 

To instruct troops in the strategy and tactics of warfare there are the 
Training Films and Film Bulletins, released on the average of one 
every week. Each of these must be considered a separate case, dis- 
tribution and print allocation depending upon a specified audience, 
upon troop strength and dispersion, and upon "play-off" time. Some 
special technical films are limited to only a dozen prints, others of 



28 ROBERT A. KISSACK, JR. Vol 46, No. l 

general nature, such as military courtesy or personal health, where 
required showings are directed by the Army, may be sent to many 
posts, camps, and stations for repeated showings to all military person- 
nel. Distribution of the Surgeon General's reconditioning films has 
become increasingly important of late and has required the establish- 
ment of new subexchanges in hospitals throughout the country. 

There are Information and Education films which were distributed 
to troops overseas long before VE Day and were held in film exchanges 
throughout the world for that victorious moment films which are 
now being shown to troops to explain demobilization and redeploy- 
ment plans and also the Army's training program for return to 
civilian life. 

The industry's gift of entertainment film to the troops overseas 
has proved to be, as everyone knows, one of the finest morale factors 
of this war. Three complete programs are sent out each week to 21 
overseas exchanges. Reports of audience reaction are regularly ob- 
tained and guide a selection board in the choice of pictures that most 
soldiers, sailors, and marines like the most. There have even been 
overseas world premieres gala performances in the fox-hole circuits 
before release of the pictures back home. 

Thus, the type of film subject, the type of audience, and other 
variable factors, determine the best distribution policy. Some 
films are planned for circuiting throughout many installations and to 
large audiences, some are sent to exchanges for spot-booking to specific 
users, and some permanently allocated for recurrent needs. When- 
ever possible, rebookings and second-run circuits, such as programs 
for transports and hospital ships, are utilized. Wherever feasible, 
mutual print loans are accomplished among the Navy, the Air Forces, 
and the Army. Always the criterion is: the fewest prints that will 
do the job adequately. Well aware of the vast footage requirements 
of the Armed Forces, the Army has a natural desire to utilize available 
footage allocation as efficiently as possible. With the complete 
statistical information reported monthly by all central exchanges 
and libraries, requests for initial distribution of new subjects and for 
reordered prints are viewed searchingly in the light of utilization 
reports. The result of these efforts has been a huge conservation of 
motion picture raw stock. 

Army Pictorial Service also provides projectors as well as films. 
Although it does not exhibit pictures to the troops, part of its function 
is to serve as a clearing house for information on equipment allocation 



Jan., 1946 ARMY FILM DISTRIBUTION AND EXHIBITION 29 

and use. In the early days there were not enough projectors to go 
around. Many and all models were pressed into use. The problem 
was particularly critical overseas where conditions were so rugged 
that projectors took an awful beating. Damp humid climates, ex- 
tremes of heat and cold, rain, sand and dust, continuous operation, 
and terrible transportation difficulties were too much for equipment 
designed for home use. Replacement parts were not only scarce, 
they were often unavailable. 

Yankee ingenuity was the only thing that made it possible for the 
show to go on. Equipment was cannibalized, new projectors and 
amplifiers assembled from parts of many others. Public address 
equipment was hooked into projector sound systems. Tail-light bulbs 
from jeeps became almost standard exciter-lamp replacements. To 
furnish a brilliant picture and adequate sound for large audiences 
a 35-mm arc projector was completely dismembered, remodeled, 
and combined with a 16-mm machine. 

Meanwhile, motion picture engineers came to the rescue. Pro- 
jection models and parts were standardized. Design was made 
sturdier for operation and for transportation under extreme condi- 
tions. Equipment was fungus-proofed. Standardized replacement 
parts began rolling off the line. Today the Army projector situation 
is in pretty fair shape. Yet much still needs to be done to increase 
the quantity, the quality, and the suitability of 16-mm projection 
equipment for changing needs. 

The distribution and use of films produced by Army Pictorial 
Service has contributed to many phases of the war effort : 

(1) Orienting troops in the background and causes of the war; 

(2~) Instructing troops in the strategy and tactics of warfare; 

(3} Fostering morale by information and entertainment films; 

(4} Study of current operations by the Army high command for future plan- 
ning; 

(5) Research and analysis of American and enemy equipment for improvement 
of equipment supplied to our troops; 

(6) Reconditioning convalescent soldiers and counseling men ready for dis- 
charge from the Army; 

(7) Informing the general public in the United States of the progress of the 
war; 

(8} Providing civilian workers with pictorial evidence of the Army's produc- 
tion requirements. 

In conclusion it must be repeated that the success of this program 
is in large measure attributable to the invaluable help and cooperation 
of the motion picture industry and the motion picture engineers of 
this country. 



AN ANALYSIS OF THE COMPARISON OF BEAM POWER AND 

TRIODE TUBES USED IN POWER AMPLIFIERS FOR 

DRIVING LOUDSPEAKERS* 



JOHN K. HILLIARD** 



* 



Summary. This paper is a description of a study of 2 types of output power 
tubes used to drive loudspeakers. Since the introduction of beam power tubes several 
years ago, engineers have not been in agreement that they gave comparable results to 
the low impedance triode type tubes. The comparison was made on 15 and 40 
amplifiers. The results indicate how both types of tubes can be made equivalent and 
inter modulation curves on all amplifiers are shown. Signal-to-noise ratios are also 
given. 

Several years ago when the beam power tube was introduced, 
designers began using this tube because of its high power output, 
high power sensitivity, and high efficiency. These outstanding fea- 
tures, however, caused several important disadvantages, the most 
important of all being its apparent high distortion when working 
into a loudspeaker. This distortion was reduced considerably, 
however, by the application of negative feedback. 

Comparisons with low-impedance triode tubes led to a widespread 
belief among a large group of engineers that the beam power tube 
could not compete in high-quality reproduction with the triode. 

This skepticism about the use of the beam power tube was ap- 
parently justified in a great many cases and as a result, tests were 
conducted to determine if the 2 types of tubes could give comparable 
results. 

The desire to use the beam power tube instead of the triode arose 
because of the advantages of high efficiency, high sensitivity, and an 
indirect heater cathode which gives less hum than the high power 
filamentary- type triode. 

Since the intermodulation method 1 of testing amplifiers appears 
to have the best correlation with actual listening tests, it was used 
for testing the 2 types of tubes. 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** Altec Lansing Corporation, Hollywood. 
30 



COMPARISON OF BEAM POWER AND TRIODE TUBES 31 



14 35 



12 




12 



14 



2 46 8 10 

OUTPUT-WATTS 
FIG. 1. Curve A 2 A3 triode amplifier; Curve B 6L6 beam power amplifier. 




FIG. 2. Beam power 15-w amplifier. 



32 



JOHN K. MILLIARD 



Vol 46, No. 1 



Early work on this comparison indicated that the output trans- 
former seemed to be the limiting factor. This resulted from the fact 
that with the use of feedback from the secondary winding, a large 
phase shift took place at both the very low and very high frequencies, 
causing "motor-boating" and supersonic high frequency singing. 
This unstability was minimized in those cases by reducing the feed- 
back in various ways. However, a comparatively large amount of 
feedback is required so that the output impedance will be low. An- 
other practice which caused severe distortion was the fallacy of using 




+ 380 V DC. 
FIG. 3. 2A3PP 10-w amplifier. 

negative feedback to correct the over-all frequency characteristics, 
and other circuit deficiencies such as excessive shunt capacity and 
small coupling condensers. 

The result of this practice was that little or no feedback was avail- 
able at the very low and very high frequencies, and accordingly the 
output impedance varied over a large ratio throughout the frequency 
band being transmitted. 

To overcome these objections an amplifier was constructed for the 
test which had the required frequency and power capacity without 
feedback. 

This was accomplished by designing an output transformer, having 



Jan., 1946 



COMPARISON OF BEAM POWER AND TRIODE TUBES 



33 



a very high self impedance, accurate balance between windings, a 
high coefficient of coupling to reduce leakage, and a very low dis- 
tributed capacity. Care was taken to determine that sufficient driver 
power to the grid of the tubes was available. With these conditions 
fulfilled intermodulation tests were made. These tests indicate 
that the distortion could be reduced to an insignificant degree up to 
a point very near the theoretical overload point. 




20 



40 



FIG. 4. 



30 

OUTPUT- WATTS 
Curve A PP-845 tubes; Curve BPP-807 tubes. 



50 



A comparison test was then set up and a critical listening group 
was invited to determine the difference, if any, between the beam 
and triode type of amplifiers. 

High-quality direct wire monitor facilities from networks, best 
available studio film, and disk records and special sound effects were 
used as the source of program material. 

One set of tests had a 6L6 push-pull amplifier in comparison with 
a 2 A3 triode amplifier for which the measured intermodulation 
products are shown in Fig. 1. The intermodulation test signals 
consist of 60 and 1000 cycles. The 1000-cycle signal is transmitted 
12 db below the 60-cycle signal. 



34 



JOHN K. MILLIARD 



Vol 46, No. 1 



Fig. 2 shows the schematic circuit of the 15-w 6L6 beam power 
tube. 

Fig. 3 shows the push-pull 2 A3 amplifier. 

The second test had an amplifier using a pair of 807 tubes in com- 
parison with a pair of 845 tubes, and their respective intermodulation 
products are shown in Fig. 4. 

Both the intermodulation products and total harmonics are plotted. 
The approximate relation between the intermodulation products and 
the total harmonics is such that the intermodulation products are 
approximately 4 times the total harmonics. 




FIG. 5. Beam power, 40-w amplifier. 

Fig. 5 shows the push-pull 807 beam power tube schematic. 

Fig. 6 shows the 87- type 50- w triode amplifier schematic. 

The feedback and output transformers were so designed that the 
measured output impedance was identical for both the beam power 
and triode amplifiers. These amplifiers were compared on a 2- way 
loudspeaker system 2 designed for high-quality, high-power repro- 
duction. 

The listening group was unanimous that the beam power tube 
amplifiers were at least equal to the triodes and some observers 
favored the beam power tubes slightly over the triode type of ampli- 
fiers. 

The reason for favoring the beam power tubes may have been a 
result of the output hum being approximately 15 db lower for the same 



Jan., 1946 



COMPARISON OF BEAM POWER AND TRIODE TUBES 35 



net gain and because the intermodulation curves indicate the lower 
intermodulation products in the average power range. 

It is well to bear in mind the fact that equalization in the feedback 
circuit to modify the frequency characteristic of an amplifier, changes 
the output impedance. This changing impedance with frequency 
can be a very undesirable condition when the amplifier is driving a 



Re 70* Ji 




ATLY USED 



50-65CYCLES 

40C WATTS MAX 
38AMPS MAX 



FIG. 6. 87-E amplifier. 



loudspeaker since the damping of the speaker depends to a degree 
upon the load impedance of the amplifier. 

The results of these tests can be summarized as follows : 

(1} Beam power tubes can deliver the same audio power as triodes with the 
same or less distortion. 

(2} A high over-all power efficiency can be obtained using relatively low plate 
voltages and inexpensive tubes. 

(3) The circuit of the beam power tubes need not be complicated. 

(4) The signal-to-noise ratio is improved. 



36 JOHN K. HILLIARD 

(5) Intel-modulation method of testing compares favorably with the listening 
tests. 

(6) Excellent output transformers are required. 

REFERENCES 

1 HILLIARD, J. K.: "Distortion Tests by the Intermodulation Method," Proc. 
I.R.E., 29, 12 (Dec., 1941), p. 614. 

2 ACADEMY OF MOTION PICTURE ARTS AND SCIENCES: "Motion Picture Sound 
Engineering," D. Van Nostrand & Co. (New York), 1938, p. 97. 



PUSH-PULL FREQUENCY MODULATED CIRCUIT AND ITS 
APPLICATION TO VIBRATORY SYSTEMS* 

ALEXIS BADMAIEFF** 



Summary. The purpose of this paper is to describe a new push-pull frequency 
modulated circuit in which the push-pull action is accomplished by varying the resonant 
frequencies of both the oscillator and the discriminator in opposite phase relation to 
each other. Modulation of the oscillator and discriminator is achieved through the 
use of 2 capacitors with a common plate and arranged in a push-pull circuit. Thus, 
if the common plate is the moving element of a vibratory system, this circuit can be used 
for measuring vibrations or for monitoring purposes. 

One of the applications of the push-pull FM circuit is to calibrate recording heads 
while cutting phonograph records. Using this calibrator, it is possible to measure, 
without mechanical or inductive coupling, the actual frequency response of the cutting 
head under normal load, and to observe the input-output characteristics as well as the 
amount of distortion. It is also possible to use it for monitoring while recording. 

1 BASIC PUSH-PULL FREQUENCY MODULATED CIRCUIT 

Considerable interest is being shown in the development of fre- 
quency modulated circuits in which the oscillator and the discrimi- 
nator are combined in one unit. This type of circuit can be used to 
convert mechanical vibrations to electrical voltage variations and is 
applicable to measuring instruments, pickup devices, and other forms 
of transducers. 1 The purpose of this paper is to discuss a new type 
of FM circuit in which the advantages of push-pull action are utilized. 

In a single-ended FM circuit, 1 when used in conjunction with a 
very small capacitor as the frequency controlling element, a nonlinear 
relation exists between the condenser plates spacing and the frequency 
which the condenser controls. The magnitude of nonlinearity de- 
pends on the ratio between the spacing of the plates and the amount 
of variation of that spacing when the plates are vibrated mechanically. 
To illustrate, consider an LC circuit, in which one condenser plate is 
moved away from the other plate during tuning. Progressively in- 
creased separation has less effect on the resonant frequency because 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** Radio Corporation of America, RCA Victor Division, Hollywood. 

37 



38 



ALEXIS BADMAIEFF 



Vol 46, No. 1 



the latter varies approximately as the square root of the distance of 
separation. This means that a nonlinear relation between the fre- 
quency and the displacement results in even harmonic distortion. 
It is well known that by combining the voltages from 2 similar non- 
linear components in any push-pull system, the distortion can be 
greatly reduced, since the even harmonic components are practically 
eliminated. 



POWDERED IRON CORE 



6 


C| ' 


_2irr. 


*z 


If- 


n^ron 


LJ 


, if , 


c, 




/ 


XV 


L| 


ULJ 


F 
P 




Pi v 

C 2 * |L 2 


^ 




If 






~r^ 






- 






_^_ 






"r" ( 






n 


(A) 

i.'--- 



POWDERED IRON CORE C 2 



6 .. 


GJ 

GI 


/lL > 


V| 
6 2 


, 1 


L| - 


u 

;c, 




O 


^ K 






I 




_J 



Cj 1.3 




FIG. 1. 04) 



Modulated oscillator FM circuit. 
FM circuit. 



+ (B) 

(B) Modulated discriminator 



The distortion owing to the nonlinear relation between condenser 
spacing and frequency, however, can be canceled in a single-sided 
FM circuit by a complementary distortion in the discriminator cir- 
cuit. This is not easy to attain, however, because the ratio of the 
spacing and the change of that spacing, which produces a given dis- 
tortion, must match inversely the complementary nonlinear slope of 
the discriminator curve. This condition can only be satisfied when 
the distortion produced by the condenser plates does not exceed the 
complementary nonlinear action of the discriminator. This indi- 
cates that the change in condenser plate spacing must be small in 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 



39 



comparison with the average spacing (in the order of 5 per cent) . To 
meet this requirement, and to produce the swing in frequency neces- 
sary to attain an appreciable amount of output, comparatively large 
plate area must be used. If, however, the plates are made small in 
comparison and their movement is large (in the order of 30 per cent) 
to secure a usable output, the second harmonic distortion produced is 
then quite large, and was measured to be about 17 per cent in one 
instance. This large amount of second harmonic content is too great 
to be canceled by the complementary nonlinear discriminator curve, 



DISCRIMINATOR 



IE 



ty 



FREQUENCY 
SHIFT 

FIG. 2. Graphical representation of the modulated oscillator FM action. 



except very close to the resonance point, where the range is liniited 
and adjustment is critical, or far off resonance where the sensitivity 
is low, as shown later in this paper under Section 2. 

Another consideration is that, in a single-sided FM circuit, only 
one side of the movable plate is active. The other is inactive since 
it is not faced by another plate. In the push-pull FM circuit, both 
sides of the common center plate are active, since each side forms one 
plate of 2 condensers. This fact reduces the effective size of the 
movable plate to a half of what is necessary in the case of a single- 
ended circuit. Further reduction in size is attained by a larger move- 
ment in a small spacing, as mentioned before, which produces an ade- 
quate frequency swing by using 30 per cent of the spacing for the 



40 



ALEXIS BADMAIEFF 



Vol 46, No. 11 



total maximum movement. These 2 factors can be advantageously 
employed in designing a translating device having a minimum mass 
and therefore contributing to a minimum mechanical impedance. 

A typical single-ended FM circuit in its elementary form is illus- 
trated in Fig. 1A. A pentode (Vi) with its cathode (K), grid (d),. 
and screen (G 2 ) is combined with the tuned circuit (Li) and a variable 
condenser (d) to form an oscillatory circuit electron-coupled 2 to 
the plate (P) and electrostatically shielded by grid (G 3 )- (which is- 
normally the suppressor element of the tube). The oscillatory 
energy from the plate is applied to a circuit which consists of another 



: 



DISCRIN INATOR 



FIG. 3. 



OSCILLATOR FREQUENCY 

Graphical representation of the modulated discriminator FM action^ 



tuned coil (L 2 ) and condenser (C 2 ). This circuit is slightly off reso- 
nance relative to the mean frequency of the oscillator and forms the 
discriminator part of the FM system. The condenser (Ci) across the 
oscillator coil (Li) is the controlling element of the FM circuit. The 
output from the discriminator tuned inductance is coupled through a 
condenser (C 3 ) and rectified by a diode (D), after which it is filtered 
of its rj component by a choke (L 3 ) and appears as a varying voltage 
across the diode load resistor (R\). The function of this system is 
graphically represented in Fig. 2, in which the oscillator mean fre- 
quency is positioned close to the discriminator's resonant peak. 
The dotted lines represent the variation of the oscillator frequency 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 



41 



and its equivalent change in the voltage amplitude of the output of 
the rectifier and filter circuits. 

Suppose the circuit just described were again used, but instead of 
modulating the oscillator, the discriminator were modulated by 
shunting a variable condenser across the resonant coil (L 2 ). As 
illustrated in Fig. IB, the frequency of the resonant peak of the dis- 
criminator would move back and forth along the frequency axis. 
The output voltage of the discriminator would be identical to that 
produced in the case in which the oscillator frequency was modulated. 




SHIFT 



FREQUENCY 



FIG. 4. 



Graphical representation of both oscillator and discriminator combined 
modulations in the push-pull FM system. 



The oscillator frequency will still cover the same part of the discrimi- 
nator peak as before. This is graphically illustrated in Fig. 3 in 
which the dotted lines represent the displacement of the discrimi- 
nator's resonant peak. If we combine both methods of modulation 
in the same circuit in such a manner that the oscillator frequency 
shifts in the opposite direction to the discriminator's resonant fre- 
quency, the voltage output from the discriminator would be twice as 
great as would be the case if only one of the factors had been varied 
through the same frequency range. If, for instance, the oscillator 
frequency moved down along the frequency axis at the same time the 
discriminator resonant peak moved up, the relative shift along the 
frequency axis would be doubled, producing twice the amplitude in 



42 



ALEXIS BADMAIEFF 



Vol 46, No. 1 



the discriminator voltage output. This is illustrated in Fig. 4. 
Since both modulations have to be in opposite phase to produce the 
push-pull action, and since a simple controlling factor is a capacity 
change in the 2 tuned circuits, the push-pull FM circuit is particularly 
adapted to the use of 2 capacitors arranged for push-pull action. 

The combining of 2 frequency modulations is accomplished by 
employing a small condenser, one side of which is connected across 
the oscillator coil and the other across the discriminator coil. The 
common center plate is grounded. If the center grounded plate is 
moved in either one of 2 directions, the frequency shifts of the oscilla- 



POWDERED IRON CORE 

8 


I 








(r 

CHOKE 


rir 1 

R l 

OUT 




6SF7 


3? 






. i/ _ 


_ 


r 


: IT 

3 J 

DISCRIMINATOR 


i U 






1. 




ii 


If 








OSCILLATOR 












4 



FIG. 5. Push-pull FM circuit combining both the oscillator and discriminator 

modulations. 

tor and the discriminator circuits will be in opposite directions to 
each other resulting in push-pull action. Such a circuit is shown in 
Fig. 5. In this circuit, coil (L\) and one half of the push-pull con- 
denser (Ci), together with the grid, cathode, and screen of the pen- 
tode-diode tube, such as a 6SF7 (Vi), form the oscillator, and the 
coil (L 2 ), the other half of the push-pull condenser (C 2 ), and the plate 
of the tube form the discriminator circuit. These 2 circuits are elec- 
tron-coupled, and shielded from each other within the tube by its 
suppressor which is grounded. The output is then rectified and 
filtered by the diode in the same tube, and the choke (L 3 ). 

It should be mentioned, at this point, that the push-pull feature 
applies only to the action of the variable capacities. No claim is 
made to any cancellation of distortion because of the nonlinearity in 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 



43 



the discriminator characteristic. Maintaining low distortion in the 
discriminator circuit depends upon the relatively small range of fre- 
quency change. The full benefit of the push-pull action in reducing 
distortion will not be realized unless the 2 parts of the system are 
closely balanced with respect to each other. To realize this, induct- 
ances of the coils must be alike and the mechanical construction of 
both sides of the push-pull capacities must be alike so as to give sub- 
stantially equal capacity, equally varied in opposite phase. 

The balance between the 2 modulated tuned circuits was checked 
by modulating the oscillator and the discriminator separately, and 
noting the outputs in each case. This was done by substituting an 
equivalent fixed capacitor for the discriminator side of the push-pull 
capacitor and with a constant modulation applied mechanically to 



COPPER PLATES, 



COMMON 




STEEL WIRE 
VIBRATORY 

CROSS-SECTION A-A 



BAKELITE 

FIG. 6. One form of push-pull FM transducer. 

the oscillator part of the push-pull capacitor, the output was then 
noted. When substituting a fixed condenser of equal value in the 
oscillator circuit and modulating the discriminator, the output was 
found to be of the same amplitude as in the case when the oscillator 
was modulated. In other words, both circuits were designed to pro- 
duce equal output voltage variations. 

In each case, however, the exact oscillator center frequency position 
on the discriminator curve was noted, and the fixed substituted ca- 
pacity adjusted so that the position was the same. This is important, 
since if the position were to change, a different amount of diode volt- 
age would be developed for a given modulation, depending on the 
slope of the discriminator curve in the section used for discrimination. 
This position was noted by observing a microammeter connected be- 
tween ground and the diode load resistor. The reading then was 
highest at the peak and proportional to the position of the oscillator 



44 



ALEXIS BADMAIEFF 



Vol 46, No. 1 



center frequency on the discriminator curve. By thus observing the 
reading, the position and the side of the discriminator curve can be 
duplicated in each case. 

Since the output voltage of the 2 sides of a push-pull system are 
additive, only half the voltage is required from each side of the system 
and, therefore, only half the frequency change to produce it. It is, 
therefore, evident that the vibratory member is capable of producing 
a large voltage amplitude with a relatively small capacity, resulting 
in a very small unit. 



f 










X 
















f 

ta 


























r 
















/ 


/ 
























JL 










)LTAGE 














/ 












g 

g 
O 










> 


/ 














O 


































/ 










































































4 




3 ; 


) 


c 


) 




i -: 


-^ 


! 





DEFLECTION IN .001 INCH 

FIG. 7. Linearity of the push-pull FM system (distortion). 
2 OPERATION 

The push-pull FM circuit, built and tested, included a push-pull 
condenser as its controlling element. This condenser consisted of a 
steel wire Vs in. long, 20 mils wide, and 8 mils thick, on each side of 
which is a plate of similar dimensions and spaced about 5 mils from 
the center movable plate as illustrated in Fig. 6. The movable steel 
wire, which is the common grounded element of the push-pull con- 
denser, is clamped at one end. Its free end can be attached to any 
mechanical system whose vibrations are to be measured, as, for in- 
stance, the stylus tip engaged in the groove of a phonograph record. 
The rigidity of this wire is low, being in the order of 1.4 X 10 6 dynes 
per centimeter. The 2 outside plates as previously described are 
connected; one across the oscillator coil, and the other across the 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 



45 



discriminator coil, with the center movable plate grounded. Using 
the circuit as described, the diode voltage was measured by a sensi- 
tive voltmeter and a curve plotted of the diode voltage versus the 
deflection of the center condenser plate. This curve is shown in 
Fig. 7. It should be noted that throughout most of its length this 
curve is practically linear. The total harmonic content represented 
by the curvature between the 2 dotted lines, A and B, amounts to 
less than one per cent. The curvature above that is caused by the non- 



190 



180 



170 



160 



110 



cr 



.9952 .9956 .9960 .9964 .9968 .9972 .9976 .9960 .9984 .9988 .9992 .9996 I.OOO 



FIG. 8. One side of the discriminator resonance curve. 



linearity of the discriminator resonance slope. When the Q of the 
discriminator coil is in the order of 200 (which is easy to attain at 
frequencies of 40 megacycles), and the mechanical vibration applied 
to the tip of the wire is == 1 mil, the output from the diode rectifier, 
as measured across the diode load resistance RI, will be in the order 
of -30 db (reference 0.006 w). 

The only nonlinear factor in this circuit is the discriminator reso- 
nance peak, one side of which is used as the slope against which the 
frequency modulation is changed to amplitude variations of the 
carrier. The calculated linearity of that slope is represented in Fig 8, 
and is based on the formula 3 



46 ALEXIS BADMAIEFF Vol 46, No. l 

+ 2.5 x W-*K* 1 



;-[a- 



as taken from an equivalent circuit shown in Fig. 8 where L and C 
are the inductance and capacity of the resonant circuit and R is the 
loss and therefore the limiting Q factor. K stands for the ratio of the 
oscillator frequency to the resonant frequency of the discriminator 
(K = f//r)' When the distortion is thus computed, the following 
table represents the per cent of second harmonic distortion as com- 
pared to the frequency swing, relative gain, and the mean frequency 
of the point on the resonance curve against which the carrier is sweep- 
ing. 

Second Harmonic in Per Cent of Relative 

K Fundamental, =*=45 KC Gain 

0.980 2.94 25.3 

0.982 3.26 28.0 

0.984 3.67 31.5 

0.986 4.20 36.0 

0.988 4.90 41.9 

0.990 5.88 48.8 

0.992 6.9 60.0 

0.994 8.0 77.5 

0.9968 6.5 123.0 

0.9976 3.2 144.5 

. 9980 Less than 0.5 156 . 

0.9981 Less than 0.5 160.0 

0.9988 12.5 180.0 



The curve is practically straight from K = 0.9972 to 0.9988. 
Thus a discriminator will cause negligible distortion if the total range 
of modulation is restricted to 80 kc for a 40-megacycle carrier. This 
is also the portion of the characteristic which provides maximum out- 
put. The range actually used covered 30 kc in each push-pull com- 
ponent, thereby effectively covering 60 kc on the discriminator curve. 

The push-pull FM circuit presents a new tool which can be applied 
to any mechanical translating or measuring device. As an example, 
in translating the mechanical vibration of a reed it is possible to place 
the 2 outside plates of the push-pull condenser on each side of the 
reed, having the reed act as the center grounded plate. In that case 
large amplitudes can be translated with minimum distortion without 
adding anything to the reed itself. 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 



47 



3 -PUSH-PULL FM CUTTER CALIBRATOR 



During recent years as the art of disk recording has become re- 
fined, a number of improvements were made in the recording head. 
Several way;? were devised to measure the ability of the cutting head 




COIL 



FIG. 



FM PLATE 



"STYLUS tfXT GROUND POTENTIAL) 
Arrangement of the FM calibrator adapted to a recorder head. 



to reproduce faithfully the sounds engraved on a disk. 4 The most 

important of these measurements is the frequency response of the 

head after compensation. This type of measurement is made in 

several ways. One of the best-known and widely used direct methods 




FIG. 10. Push-pull FM plate assembly. 

is by the light pattern. As is generally known, this method consists 
of recording several frequencies on a disk and comparing their am- 
plitudes with the aid of reflection of light. This method is fairly ac- 
curate when certain precautions are taken and is really a true indica- 



48 ALEXIS BAD^AIEFF Vol 46, No. 1 

tion of the performance of the head, since the head is actually engaged 
in cutting a disk that represents the normal load on the stylus. 

Another method which has been widely used consists in measuring 
the amplitude of the stylus vibrating freely in air by means of a mi- 
croscope which has been precalibrated to measure small distances. 
In this method it is very difficult, however, to obtain accurate results, 
since in some cases the amplitude of the vibrations is extremely small. 
By substituting a photoelectric cell for the human eye it is possible to 
improve the accuracy to a large extent. Such optical cutter calibra- 
tors have been in use for years. Since the measurements are taken 




FIG. 11. Component parts of a complete FM calibrator for recording heads. 

in air, it is assumed that the difference in the measurements is small 
when compared to such made while the head is actually cutting a 
record. 

The problem of being able to calibrate the recorder under actual 
cutting conditions still remained, however, and it was not until the 
push-pull FM circuit was developed that a solution to the problem 
was realized. A device was needed which could be attached to the 
recorder without requiring much space or adding any mass to the 
moving system, as well as one which would not couple electrically 
to the driving coil of the recording head, and of course one which 
would not interfere with the cutting action of the stylus. 

The push-pull FM system meets these requirements very nicely. 
However, it is practically essential to use the push-pull circuit, be- 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED ClRCUIT 



49 



cause the distortion would be too high if only one side of the stylus 
were used as one of the plates of a condenser, owing to the low ratio 
of the capacitor plate spacing and the large variation of the spacing. 
As illustrated in Fig. 9, 2 small plates, one on each side of the stylus 
shank, insulated from each other and from the recorder, are spaced a 
few thousandths of an inch from the stylus. Since neither mass nor 
rigidity is added to the moving system, no change in its action can 
occur. Two flexible leads from these plates and another lead from 
the cutter mechanism are connected to the oscillator discriminator 
unit mounted on the carriage and located as close to the cutter head 
as feasible. The oscillator-discriminator unit is of the same type as 




FIG. 12. The FM calibrator in operation. 

described in Section 1. The variation of the capacitance, between 
the 2 plates and the stylus, because of its motion, changes the oscilla- 
tor frequency and varies the tuning of the discriminator in accordance 
with the method of operation described in Section 2. The audio out- 
put of the oscillator-discriminator unit, since it also contains a diode, 
is transferred to another unit containing an amplifier and a power 
supply. The output from this amplifier is then either measured with 
the aid of a tube voltmeter, or is further amplified for listening or 
other purposes by using a suitable transformer located between the 
amplifier tube, which has a high-impedance output, and a conven- 
tional amplifier of 500 ohms input impedance. 

The push-pull FM plates are mounted on the bottom side of the 
cutter in such a way that in combination with the cutting stylus they 
form a small push-pull condenser. As shown in Fig. 10, this assem- 



50 



ALEXIS BADMAIEFF 



Vol 46, No. 1 



bled bakelite plate on which the 2 FM plates are mounted fits on the 
under side of the cutter and replaces the cover plate. This assembly 
is again shown separately in Fig. 11. The FM plates are, of course, 
adjustable so as to provide any width of gap to accommodate any 
cutting styli or for unusual circumstances where a wide gap is neces- 
sary. Normally, however, the FM plates are spaced 0.01 in. away 
from the stylus for frequency response measurements and 0.015 in. 
for distortion measurements at the lower frequencies. This spacing 
is ample for any modulation and yet provides more than ample out- 
put. 




FIG. 13. Schematic of the push-pull FM calibrator. 

For calibration during the cutting of a record, the cutter is mounted 
in the usual way. The oscillator-discriminator unit is mounted im- 
mediately beside the cutter (see Fig. 12). The FM plates are con- 
nected to the oscillator and discriminator circuits by means of two 
0.004-in. diameter steel wires covered with vinylite insulation. The 
ground wire between the 2 units is of the same material. The com- 
plete schematic diagram of the calibrator is shown in Fig. 13. 

The output of the oscillator-discriminator unit is fed to an amplifier, 
which is on the same chassis with the power supply. It consists of a 
conventional circuit using the 6SJ7 triode connected. From there 



Jan., 1946 PUSH-PULL FREQUENCY MODULATED CIRCUIT 51 

the signal passes through a 3-position switch which selects the "flat 
compensator," the "orthacoustic," or the "no compensation" circuits. 
These positions can be used for different purposes as follows: "No 
compensation" is used for measurements with a vacuum tube volt- 
meter, for frequency response tests, and input versus output curves. 
"Flat" response can be used for monitoring on disks requiring that 
type of response; however, the important function of this position 
is for intermodulation tests. "Orthacoustic" can also be used for 
monitoring while the record is being cut, and is designed for that 
purpose. 

To obtain consistent readings, it is important to have the oscillator 
tuned to the same frequency relative to the optimum point of opera- 
tion of the discriminator circuit. To accomplish this, a meter is pro- 
vided in the diode circuit. The oscillator circuit is tuned for maxi- 
mum diode current and is then backed down, in the same direction 
each time, to 70 per cent of the peak current. 

The use of this instrument and results of measurements are com- 
pletely covered in the paper by H. E. Roys entitled "Experience with 
an FM Calibrator for Disk Recording Heads." 5 

ACKNOWLEDGMENT 

The author wishes to acknowledge the helpful suggestions and ad- 
vice of E. W. Kellogg, H. E. Roys, C. M. Sinnett, and A. C. Blaney 
in this work. 

REFERENCES 

1 BEERS, G. L., AND SINNETT, C. M. : "Some Recent Developments in Record 
Reproducing Systems," /. Soc. Mot. Pict. Eng., XL, 4 (Apr., 1943), p. 222. 

2 TERMAN, F. E. : "Radio Engineering," McGraw-Hill (New York), 1937, p. 360; 
Dow, J. B.: "A Recent Development in Vacuum Tube Oscillator Circuits," 
Proc. I. R. ., 19 (Dec., 1931), p. 2095. 

3 LANDON, V. D., AND MC!LWAIN, K.: "Fender's and Mcllwain's Electrical 
Engineers' Handbook (Electric Communication and Electronics)," John Wiley 
and Sons (New York), 3d Ed., p. 7. 

4 FINK, D. G.: "Electronic's Engineering Manual," McGraw-Hill (New York), 
p. 240. 

6 ROYS, H. E. : "Experience with an FM Calibrator for Disk Recording Heads," 
J. Soc. Mot. Pict. Eng., 44, 6 (June, 1945), p. 461. 



A DISCUSSION OF THE ACOUSTICAL PROPERTIES OF 

FIBERGLAS* 



WILLIS M. REES AND ROBERT B. TAYLOR** 



Summary. The wide range and controllable properties of Fiber glas thermal 
insulating materials when used as absorbents of airborne sounds are discussed. 
Tables of sound absorption coefficients are given for 6 frequencies from 128 to 4096 
cps and for 24 different combinations, weights, and thicknesses of materials. Curves 
are shown analyzing the relative absorption at different frequencies of the different 
materials based on density, thickness, and fiber size. 

Sound isolation blankets made of a newly developed superfine glass fiber show re- 
markable acoustical efficiency when used for aircraft sound insulations. Their 
potential application to other uses when materials and research facilities are available 
is discussed. Some data are also given evaluating the influence of surface treatments 
and enclosing materials and methods. 

The paper is a compilation and analysis of available data but does not represent 
a complete evaluation since much further work is needed as a result of properties 
discovered in war applications. 

During the 10 years prior to the war, glass products in fibrous 
form, known as Fiberglas, were being developed from those of broom- 
straw diameters to those of microscopic fineness. With each refine- 
ment in manufacturing processes and development of finer fibered 
products, entirely new groups of products and applications for them 
came into being. 

The original type of large diameter fibers are used in air filters 
and aeration mats. The next group of products of much smaller 
fiber diameter are known as wool products and are used for thermal 
insulation and sound absorption. Smaller diameter fibers were 
developed having microscopic fineness and great length. These 
became textile materials which are used for electrical insulations, for 
fireproof fabrics and draperies, for plastic reinforcements, and a host 
of other applications. The most recent development is down-like 
fiber of submicroscopic fineness which is made into incombustible 
aircraft blankets with exceptional heat and sound insulating proper- 

* Presented May 18, 1945, at the Technical Conference in Hollywood. 
** Owens-Corning Fiberglas Corporation, Toledo, Ohio. 

52 



ACOUSTICAL PROPERTIES OF FIBERGLAS 



53 



ties, combined with extremely low weight. About 8 standard fibers 
are being produced ranging in diameter from 0.008 in. for aeration 
products down to 0.00022 in. for textiles, and 0.00005 in. for the new 
aircraft blankets. Basic characteristics of various fibers are shown 
in Figs. 1 and 2. 

During the prewar period acoustical tests were made at Riverbank 
Laboratories of various Fiberglas wool products. Some of the test 
results are here made public for the first time. Our wish is to give 
you a preliminary report, or a preview of what we believe may lie 



1.000,000 
900.000 



TENSILE STRENGTH OF FIBERGLAS BASIC FIBERS ^L 
IN RELATION ID FIBER *"""^ 




FIBER DIAMETER IN HUNDRED THOUSANDTHS OF AN INCH 



FIG. 1. 

ahead in the acoustical development of Fiberglas materials based 
on a critical analysis of these data and further tests which have 
been made in the Fiberglas Research Laboratories. However, it should 
be pointed out that there are not enough data on which to base any 
scientific conclusions. The data are considered merely as a guide to 
point the direction for further study and research. 

The plain white insulating wool form of Fiberglas is known as 
"TW-F Wool." It is manufactured to a natural density of l l / 2 lb 
per cu ft but because of its resilience it may be compressed to greater 
density. It is thus possible to use the material in a wide range of 
densities and thicknesses. The same material, when bound with a 
thermosetting resin, forms a serie^ of preformed products known as 



WILLIS M. REES AND ROBERT B. TAILOR Vol 46, 



(M <M (M <M (M (M 



rt 

OJ OJ 



O W53 

? .c6 



tj i C "Sj "S "5 

TO ' - 1 



15,5,35 



> rt 



3 "a s ^ s """a i ^ "3 "3 fi "3 s iSs 5-s 

JQJ ^j rtj L.J ^j &) uj w^ * < t < rt ^H Ofl rt Ctf v* rt Ctf 

^2n^r\^3^n^^>^S rX ^>. n.n.o. m nn m r * A m 



pQPQPQOd. PQ 



a ^ 

^ _E5 



IO t (-rTCNCOCNC^JCOUUl-^-t I i IT I lOt^l^- 

CD O CD *O CD CD O CD CD CD ' i ' O CO CO 

<M co <M co <M <M co <M oq <N c<j M oq co co co 



CO ^ CO O CO 1C 



ij 



b- I> O CN 

CD CD O "^f 



O 
iO 



'-i iO O O iO ^ 
(M CO 1C to 1^ t^ 





5, 


^ 


^4 


1 


O <U ^ 

^2^ 


^ 


^ 


'" 1 


^4 


g 


a 


^ 


fe 








a 

g 


o 

(t- 


^c 


( 


> 


1 


.^H -co -CO -t^ COOS 

.t->. -t> . -oo -oo r^-io 


OO<N <oi-iO3i-Ht--<MiO 

000000 8000000t^OOO 


H 


s 


1 


t 


L 

xo 


^ 




ft, 




1 




,1 


s 

I 


GC 

i 


OCOOO!>t^COt^OOCOT-ii-i lOCO 


OS CD i i ^ COCO'^TtiOO't 

i>ooos t^oooooocooo 




1 


t/3 
1 

2 


i 


| 

! 

H 


O 


OS t^ 00 r-i CO 1-1 CO OS OS OS O ># OS 
co >. QQ as as OS OS OS Os Os OS OS OS 


Oi OS OS t>- 00 OS as OS OS 




1 


1 




"S 

1 


(N 


fs^. QQ ,__) fQ ^ ^ ^Q Q} ^> OS CO 00 OS 


tOOsOs totOcDtot^-OS 

as as as CD r^ oo r^ as as 




cl 


1 




1 












1 






i 


010*001 lOSr- tOOtOOsO Os^f 
CO CO "* Tj< tO "* CD CD 00 l>- CD tOCO 


00 00 i i (M i i 1-1 CO CD O 
CD l^ OS CO Tf Tf CO Tfi CD 






.> 
















W 






oo 

(M 


T^t^COCOtOOOrf^(MOO 001^ 
Cq<MCOCOCOCO^iOCDiOLO T^T^ 


toooas t o o oo c^i i-t 






i 






a_ 


iO *O *-O ! O O *-O ! O O O 1 O 1 O ^O ^O 
iO CO t^* t^* 00 t^ 00 O^ O^ O5 00 00 00 


lOOtO OOtOOiOtO 
OOOsOs CDC^-t^t^-t^OO 






?? 






^ 







<M ^ CD CD OS Tt< 



<N (N CO CO CO 



Jan., 1946 



ACOUSTICAL PROPERTIES OF FIBERGLAS 



55 



rfa 2 

OJ J- 

"o 3 ^ 

& 1 

oo .S <y 

I| I 

T3 CT bO 

C i _C 

rt t/i *7"! 



|| Sfe 



Oo. 



CO O 

^H 1C 



CO t^ 

OS Os 



O OS 
1^- OS 



T-H CO 



rt S 

w ^ 

'ti ^4 

?, o 



OJ 
O OJ ' 

aj S -2 



II 



:S3 rg s 



x 



43 rt 

rt C 






PQ 
^g 



a3oJajajQJaJ <u *S 

a a a, cx a a a <j 



_, osoo^ost^co'-'aj 

I 832S3S 




56 



WILLIS M. REES AND ROBERT B. TAYLOR Vol 46, No. 1 



"PF" materials made in flexible or rigid block or board form. These 
can be made in predetermined thicknesses and densities, the latter 
ranging from 2 x /2 lb to 9 Ib per cu ft. They are also incombustible 
and chemically inert, as are all Fiberglas wool products. 

Table 1 gives sound absorption data for type TW-F Wool in several 
densities and thicknesses. All items show exceptionally high sound 

SURFACE AREAS OF FIBERGLAS 
BASIC FIBERS 

( PER POUND Of 01 



*AA" FIBERS - 6OOO SQ. FT. 




FIG. 2. 



absorption at the low frequencies. Some show coefficients at 128 
cycles considerably in excess of 0.50 and even as high as 0.69. All 
but one exceeds 0.25, a characteristic possessed by very few of the 
usual acoustical materials. 

Table 1 also gives data for several PF densities and thicknesses. 
The same general acoustical characteristics of high absorption at low 
frequencies are evident here (as with TW-F type of materials) al- 
though to a lesser degree. 



Jan., 1946 



ACOUSTICAL PROPERTIES OF FIBERGLAS 



57 



All of the data given in Table 1 are from tests made by the River- 
bank Laboratories by methods prescribed by the Acoustical Ma- 
terials Association. 

Fig. 3 shows the apparent effect of thickness and density upon ab- 
sorption values of TW-F Wool. In a material one inch thick, the 
absorption at all frequencies increases with little change in the shape 
of the curve as the density increases. As no tests have been made for 
densities over 6 Ib, it is not apparent what the optimum density would 



J..UU 

.ao 

,60 
,40 




^ 


^^~ 





"^^ 


r^^. 


jjt 

ckn 


*<*. 


/ 










^ 


/ 








3" thi 














1.00 

.80 
.60 
.40 














ckn 






/<&. 


^~~^ 


^T^ 


4* ^ 


f* 


/^ 


-/'Y 






~~1* 


1 


^ ^ 


7 






2" thi 


z* 












l.OU 

.80 
.60 

.40 
.20 







^ 


f ~^_. 


^^ * 


!* 


ckn 






/<''A 


^~^~- 





$--* 


ft 


^ 


''',S/ 






1" thi 


; 


"" ^ ** 


** 








2? 















128 256 512 1024 2049 
TW-F WOOL - THICKNESS vs DENSITY 

FIG. 3. 



4096 



be for one-inch material. The 2-in. thickness curves show increased 
absorption at all frequencies with little advantage of 6-lb material 
over 4-lb material at middle and high frequencies. In the 3-in. 
thickness curves it would appear that density is of little consequence 
for frequencies of 5.12 and up, although higher densities improve low- 
frequency values. 

These curves also show that an engineer who is concerned with 
the economics of an acoustical installation has a considerable choice 
of materials. If he is concerned only with higher frequencies, a 
2-lb density material 3 in. thick is just as good as double the amount 
of material in a 4-lb density 3-in. thickness. 



58 



WILLIS M. REES AND ROBERT B. TAYLOR Vol 46, No. l 



Although the data are not complete enough to make positive asser- 
tions, it would appear that for all practical applications of 2-in. or 
3-in. thickness that a 4-lb density of TW-F Wool is optimum density. 
In the 3-in. thickness, further tests may show that greater than 4-lb 
densities will increase low-frequency values giving a high flat curve. 
This appears to be a logical deduction from curves shown and in- 
dicates a possible course of investigation. 



i.QO 
.80 

.60 
.40 



E 



1.00 



,do 



,bQ 



.40 



23". 
3**' 



x 



1.00 Ibs 



per sq 



ft. 



,50 Ibs. per sq. ft 



z 1,00 

.80 

^ .60 

.40 

.20 



.33 Ibs. per sq. 



ft 



**/ 



128 256 512 1024 2048 4096 

FREQUENCY 
TW-F WOOL - WEIGOT vs THICKNESS 

FIG. 4. 



It has been said that friction of pulsating air molecules against the 
walls of the interstices of porous materials dissipates sound energy 
into heat. It is in this way that most absorption is provided. Fig. 2 
shows that there is a great increase in the square foot surface area 
of a pound of Fiberglas as the fiber diameter decreases. Because 
surface area within a given weight of Fiberglas is related to the di- 
ameter of the fibers, research may show that acoustical values are 
functions of the fiber diameter. Test work in a war research sound 
laboratory indicates that this may be true. 

In a fibrous product such as Fiberglas TW-F Wool, it is conceivable 
that within reasonable limitations this frictional loss would be a func- 



Jan., 1946 



ACOUSTICAL PROPERTIES OF FIBERGLAS 



59 



tion of the surface area of the fibers which make up the material. 
If such a supposition were true, it might be just as reasonable to be- 
lieve that a given square foot weight of material with given surface 
area of fiber would have about the same acoustical value whether it 
was applied thick or thin, dense or light. If surface area is what 
does the trick, pounds of a given type of fiber and not thickness or 
density should be the controlling factor. 

In Fig. 4 are plotted the few data available to show the effect of 
applying a given weight of TW-F Wool in several thicknesses and 
densities. Admittedly, there are insufficient data to permit positive 



.30 




2 


^ 


= ^^^ 


^-^: 


one coat casein paint 
plain bare board 






2 


~~~"-- 


^^- 





two coats casein paint 






/ 






"^~- 


one coat lead and oil 
paint 











































~*~ 








^^ 


^=^ 










\ // 


s 




\f 


-loose Cellophane wrapped 






JzGfc 


^.x-"' 


'"-.^ 


^,\ 


Cellophane cemented to 
surface 




* 


^*\> 


^=^=^ 


^^ 


^s. 


two coats lead and oil 


T) 




"^ 


^.^s 


^s? 


^ 


three coats lead and oil 
paint 



FREQUENCY 
APPABSWT EFFECT OF SURFACE FILMS 

FIG. 5. 



conclusions. But there is surprisingly little difference for variations 
in the density and thickness of application of a given square foot 
weight of material. It does appear, however, that if an acoustical 
engineer were given a choice of thicknesses in which to install a given 
weight of material, the results would favor the greater thickness. 

To study the effect of surface treatments on a Fiberglas board prod- 
uct of the PF type, acoustical tests were made in the Fiberglas Re- 
search Laboratories. These tests were made by the tube method, 
measuring sound attenuation in a concrete tube lined with the ma- 
terial under test. The term "Absorption Rating" is used in presenting 
these data rather than "Absorption Coefficient" because the informa- 
tion given provides comparative data only and is not intended for 



60 WILLIS M. REES AND ROBERT B. TAYLOR Vol 46, No. l 

direct comparison with data obtained by the reverberation chamber 
method. 

Fig. 5 shows the apparent effect of various surface films on a 
Fiberglas board. The curve for bare board is typical of many ma- 
terials. The upper set of curves shows that one coat of casein paint 
has no appreciable affect on the absorption. But with either 2 coats 
of casein paint or one coat of lead and oil, strange things happen. 
The curve is completely altered, both surfaces giving substantially 
the same characteristics. The test with 2 coats of casein in par- 
ticular gives an extremely flat curve. 

In the lower set of curves in Fig. 5, the effect of additional paint 
coats is shown. Two and 3 coats of lead and oil, and 4 coats of casein 
all show very similar curves. Cellophane cemented to the board 
surface and even solid unperforated 26-gage sheet steel surfacings 
show similar curves. A loose Cellophane wrapping shows little 
change in the curve from the bare board except for a reduction at 
4000 cycles. 

Interpretation of the curves in Fig. 5 is not as simple or obvious 
as those for TW-F Wool in Figs. 3 and 4. Fiberglas board is fairly 
rigid but has some springiness or compressibility. A film which 
sealed the surface could also be subject to diaphragmatic action 
owing to resilience of the board. It is believed that the heavy paint 
films and cemented Cellophane have one thing in common; namely, 
that they seal the surface porosity but still permit a movement and 
dissipation of sound energy owing to resilience of the board. It is 
believed that the single lead and oil paint coat and the double casein 
coat have similar curves because of similarity of restricted surface 
porosity, possibly as a single characteristic or in combination with the 
property of resilience. In any event, the curve for 2 coats of casein 
paint is one of extreme flatness which might be of great value for 
certain critical applications if it could be designed into a product. 

What effect varying degrees of resilience in Fiberglas boards might 
have on the absorption curves with sealed surfaces is a question. 
But resilience might be another controllable factor in the design of 
an acoustical product, although this characteristic in Fiberglas board 
is less subject to complete control than some of the other traits dis- 
cussed. 

Fig. 6 shows the effect of perforated metal facings with different 
percentages of open area. With a high percentage of open area, such 
as 18 per cent, there appears to be little effect upon the absorption 



Jan., 1946 



ACOUSTICAL PROPERTIES OF FIBERGLAS 



61 



curve except for slight reduction at 4000 cycles. As the percentage 
of open area is reduced, high frequency values decrease and low fre- 
quency values increase. Reducing the percentage of open area seems 
to sharpen the peak of the curve and move the peak to lower fre- 
quencies. 

No study will be complete, however, until the relationship between 
acoustical effect upon absorbents covered and hole size and spacing 
for various open areas is investigated. It will also be important to 
study the effect of various types of perforated surfaces over materials 
of varying characteristics. 

During analysis of the acoustical data on Fiberglas materials and 
limited research findings, one fact stands out: Fiberglas materials 



plain bare board 



l8 open area 
\/k" holes - 1/2" o.e. 



k.<)f. open area 
l/k" holes - 1" o.c. 



'.2 open area 

1/8- holes - 1" o.c. 



0% open area 
not perforated 



ABSORPTION RATING 

8 k k g g 




^. 


^<z^^^ 


^^rr: 






/ / 

/// 


^ 7 


x X 

\ \ 


\ 

\ N N 




1 




\ 
\ 


"X. 

<" 




\. 






"- - 




X K 


-v-*""*" 


"\ 





250 



500 1000 

FREQUENCY 



APPARENT EFFECT OF PERCENTAGE OPES AREA 
IN 26 OA . PERFORATED METAL SURFACES 



FIG. 6. 



provide a group of incombustible products apparently subject to 
more complete control than has ever before been possible in acoustical 
materials. With control of thickness, density, fiber diameter, and 
surface area per unit of volume or weight; with possible control of 
resilience, porosity of surface films, combinations with perforated 
surfaces; and with control of methods of mounting and assembling 
materials in actual use, the conclusion is warranted that research men 
can find in Fiberglas materials the means of conducting basic research 
into the physics of sound absorption. 

Through research there may be developed some day products 
that will more closely meet the requirements of the motion picture, 
radio, and television industries. As more is learned about the needs 
of these industries and of the acoustical characteristics of Fiberglas 
materials and the methods of using them, the availability of ma- 
terials to meet specific requirements may become a reality. 



62 WILLIS M. REES AND ROBERT B. TAYLOR Vol 46, No. i 

There are other acoustical applications for Fiberglas materials 
than as sound absorbents. Fiberglas Bonded Mat is used as a 
covering material over either TW-F or PF types of products. It is 
a porous, meshed sheet formed by bonding together long glass fibers. 
It is made in thicknesses from 0.010 in. to 0.050 in. The 0.010-in. 
thickness is usually used as an acoustical facing as it offers desired 
characteristics at minimum cost. It may be sprayed with casein 
paint without apparently affecting sound transparency. 

So far as is known glass cloth draperies have not been tested for 
their acoustical absorbing values. However, some of the lighter 
weight cloths have been used as facings on sewn blankets with a 
Fiberglas wool filler. The high acoustical value for such blankets 
would indicate sound transparency for the cloth, permitting its rather 
free use over sound absorbents. 

During the war the speed, power, and size of all types of aircraft 
have been tremendously increased. These changes have resulted in 
more difficult sound control problems and have rendered inadequate 
the types of acoustical treatment previously used. 

A great deal of money and effort was expended by the National De- 
fense Research Committee at Cruft Laboratory, Harvard Univer- 
sity, to develop improved types of aircraft acoustical materials and 
treatments. Since this work was primarily directed at the aircraft 
field, low weight was an important design consideration. Also it was 
recognized early in the program that it would be impossible to develop 
a light-weight structure offering much attenuation to the very low 
frequencies, so the evaluation work was principally confined to the 
frequency range of 1650-9100 cycles per sec. This is the portion of 
the audio spectrum in which noise interferes most with aural com- 
munication. 

This research effort has resulted in the development of several very 
efficient structures for the above frequency range. Coinciding with 
this structure research, 2 new types of Fiberglas were developed: 
XM-PF and XAA-PF. Both materials are admirably suited to such 
light-weight applications, possess extremely good acoustical proper- 
ties, and have the additional advantages of inherent fireproofness 
and low moisture absorption. 

Type XM-PF blankets are approximately equal to Kapok acous- 
tically when compared on a weight basis. Type XAA-PF is a su- 
perior grade Fiberglas product and will do an equally good acoustical 
job at weights 1 / 9 to I* that required with Kapok. Using this ma- 



Jan., 1946 ACOUSTICAL PROPERTIES OF FlBERGLAS 63 

terial, blanket weights of 2 J /2 oz per sq ft will satisfactorily insulate 
multimotored heavy bombers. Such a blanket is so light one has a 
difficult time realizing its acoustical efficiency. 

The full details of these new acoustical treatments are still re- 
stricted. However, they should find many applications in the post- 
war period in mobile equipment, portable sound studios, light-weight 
sound isolating blankets, etc. 

Fiberglas research men are convinced that with proper engineering 
and use of principles that have been incorporated in aircraft design, 
important contributions will be made some day to other industries 
having problems of noise control that must be solved by isolation 
of airborne sounds. 



A THREE-BAND VARIABLE EQUALIZER* 
L. D. GRIGNON** 



Summary. A recently designed variable equalizer providing suppression and 
emphasis in 3 frequency bands, adjustable by means of 3 controls, and based on a 
different philosophy, is described. The principle features include, (1} zero insertion 
loss, (2} small change in apparent insertion loss as equalization is varied, and (5) 
more pronounced effects for equivalent changes in equalization. 

Since the commercialization of sound motion pictures, the tech- 
nician has been designing, modifying, or discarding equalizer net- 
works having various frequency characteristics and configurations to 
such a prodigious extent that a list of the literature would be exceed- 
ingly lengthy. Let it be sufficient to note that the most recent in- 
formation on the subject was presented by Miller and Kimball, 1 
which described a system of networks so designed as to cover the 
frequency spectrum in discrete bands and with sufficient flexibility 
as to provide a great variety of choices to the user. This paper de- 
scribes another system of corrective networks based on a differing 
philosophy with the end result achieved in an unusual manner. 

The variety of corrective networks in use results from the equally 
great number, and the extent, of the required corrections encountered 
over a period of time and to the difficulty of adequately and rapidly 
analyzing undesirable features in the signal being recorded or repro- 
duced. We must, therefore, decide how far we wish to go in provid- 
ing the corrective devices and what frequency characteristics are to 
be used ; particularly, if the desire is to minimize the total number of 
units to be supplied. The bases for the specifications of the device to 
be described are as follows : 

(I) The majority of the desired corrections exist in either the low or high ends 
of the present-day frequency range or both simultaneously. 

(2} Intelligibility must be maintained at all times, therefore, a means must 
be provided to emphasize the mid-spectrum frequencies. Conversely, suppression 

* Presented May 15, 1945, at the Technical Conference in Hollywood. 
** Twentieth Century-Fox Film Corporation, Beverly Hills, Calif. 

64 



A THREE-BAND VARIABLE EQUALIZER 



65 



of the same band can frequently be used to make harshness more tolerable, pro- 
vided intelligibility is not degraded. 

(3) The number of units shall be kept to a minimum and still provide the 
greatest amount of correction. 

Accepting the above fundamentals as design features, the complete 
specifications can be written as follows : 

(1} Low- and high-end emphasis or suppression shall be in discrete steps and 
achieved by a shelf effect, arranged to shift the frequency of one-half loss or gain 
further toward the extremes of the signal band or nearer the mid-frequencies, 
thereby affecting the normal circuit characteristics only to the extent required by 



LOW FIEQIENIII 



HIGH FREQ 



JECCIE 



FIG. 1. Low- and high-end equalization specifications. 



the necessary correction. (This differs from the usual manner in which only the 
slopes of the characteristic are changed, the "hinge" frequency remaining fixed.) 
Maximum suppression shall be approximately 6 db, maximum emphasis approxi- 
mately 8 db, and change between steps approximately 1 db, measured at 100 
cps and 7000 cps. Fig. 1 gives these requirements in more detail. 

(2} Mid-frequency correction shall have a maximum at 2750 cps in an amount 
equal to 5 db suppression, or emphasis, in steps not exceeding x /2 db (Fig. 2). 

(3) The insertion loss of the system of networks shall not exceed 1 db at 
normal settings. 

(4} Suppression, at any of the 3 bands, shall be introduced by rotating a 
control in a counterclockwise direction from a normal setting and emphasis shall 
be applied by rotating the same control from the same normal setting in a clockwise 
direction. In either case, suppression or emphasis is made progressively greater 
as the control is manipulated further from the normal. 



66 



L. D. GRIGNON 



Vol 46, No. 1 



(5) As suppression or emphasis is introduced into any of the 3 frequency 
bands, the insertion loss at 1000 cps shall not vary more than =t 1 db. 

(6) Dial stops, which are readily adjustable, shall be provided on each control 
to permit pre-setting, or to provide limits. 

(7) The complete system shall be constructed on a portable plug-in basis. 

The above specifications assure the condition that the system may 
be inserted into a circuit without affecting normal transmission, ma- 
nipulations, or usage and that volume corrections need not necessarily 
be made simultaneously as equalization is introduced. Under some 
circumstances, particularly when low-end corrections are applied, 



FREQUBNC IE 



FIG. 2. Mid -frequency specifications. 

there will be an apparent level change which this specification does 
not consider. A portable type of construction was selected since it 
is not proposed to build any final systems until certain special items 
of apparatus are more easily obtainable in the post-war period. In 
the meantime, a small number of systems, available wherever re- 
quired, are expected to serve the immediate needs. 

The mid-frequency band arrangement will be described first. 

MID-FREQUENCY SECTION 

Two constant impedance networks are used for this control; one, 
a suppression network of the required shape and fixed in amount at 
the desired maximum; the second, an emphasis network connected 



Jan., 1946 



A THREE-BAND VARIABLE EQUALIZER 



67 



in series and variable in equalization by an amount equal to the sum 
of the maximum amounts of desired suppression and emphasis. At 
an intermediate point in the setting of the variable attenuator as- 
sociated with the variable network, the frequency characteristics of 
the 2 networks are inverse and the net applied frequency correction 
i^ zero. The design parameters for the particular specifications are 



F r 



= 2750 cps 
10 



Z = 500 ohms 

Pad loss = 5 db 



Pr 

b 

Z 

Pad loss (design value) 

Pad loss (actual) 



= 2750 cps 

10 

= 7.5 

= 500 ohms 
= 8db 
= to 10 db 





MID-FREQUENCY EQUALIZER 
FIG. 3. Mid-frequency circuit arrangement. 

In order to satisfy (5) of the specifications a second variable at- 
tenuator having a 10 db maximum was placed on the same shaft with 
the variable network attenuator, connected in series with the 2 net- 
works, and mechanically arranged to maintain the total attenuator 
loss at 10 db. 

The arrangement and circuit as described above is shown in Fig. 3. 

LOW- AND HIGH-END SECTIONS 

Having already introduced an insertion loss of 10 db (owing to the 
mid-frequency networks) it becomes necessary to use at least 10 db 



68 



L. D. GRIGNON 



Yol 46, Xo. 1 



of gainhn order to satisfy (3) of the specifications. By providing an 
amplifier of greater than 10 db of gain, the remaining requirements can 
be realized by use of a negative feedback path. The theory of am- 
plifier negative feedback is widely known and will not be discussed 
fully in this paper. Excellent references are provided by Black 2 and 
Terman. 3 

The fundamental equation pertaining to negative feedback is 



Net gain 



1 - KA 



where A = voltage amplification 

K = ratio of applied input feedback voltage to the output voltage. 




V2 



i 



FIG. 4. Typical 2-stage amplifier with parallel feedback. 



Examination of this equation discloses that a two-fold change in 
factor K will provide 6 db change in gain, provided the product KA 
is made sufficiently large. Further, by selection of the proper value 
for KA the slope of the gain change can be adjusted up to a maximum 
of 6 db per octave. If K is caused to vary with frequency by the 
use of reactive elements, the phase angle around the feedback loop 
must be considered and the calculations for net gain become much 
more laborious. 

Fig. 4 depicts a typical 2-stage amplifier with parallel feedback from 
the second plate circuit to the input cathode resistor. With fixed 
input and output voltages assumed, the feedback factor is determined 
by the ratio of Zi to Zi + Z 2 and if either Zi or Z 2 , or both, are made a 
function of the applied frequency, the characteristic of the amplifier 
will vary in accordance with the given equation. 

Low- and high-end suppression, or emphasis, as required in this 



Jan., 1946 



A THREE-BAND VARIABLE EQUALIZER 



69 



case can be the result of simple condenser-resistor combinations such 
as are shown in Fig. 5 A. With a suitable choice of resistors, ca- 
pacitors, over-all gain (without feedback), and feedback factor, the 
specified frequency characteristics can be obtained by increasing or 
decreasing C\, Cz, 3, or C 4 in discrete steps, as follows : 

Low-end suppression: Decrease C\ 

Low-end emphasis: Decrease 3 

High-end suppression : Increase C 4 

High-end emphasis: Increase Ca 





FEEDBACK NETWORK 



EQUIVALENT NETWORK 

AT 
ZERO EQUALIZATIONS 



FIG. 5. (^4) tylethod of connecting capacitors and resistors in feedback 
path, and (B) equivalent feedback path at zero equalizations. 



Capacitors C\ and Ca are connected into the circuit with a single 
switch which is arranged to maintain C\ constant in value, while 3 
is being decreased to produce low-end emphasis and, conversely, C 3 
is held constant as C\ is decreased to give low-end suppression. Ca- 
pacitors Ci and C\ are switched in the same manner except that either 
is held at the minimum value while the other is increased, in steps, to 
obtain high-end emphasis or suppression. The approximate feed- 
back network, for zero equalizations, is shown in Fig. 5B. 

It has already been indicated that large values of KA must gener- 
ally be used which can obtain by a variety of choices for the separate 
values of K and A. In order that the feedback path does not 
adversely shunt down the plate cathode branch of the output tube, 



70 



L. D. GRIGNON 



Vol 46, No. 1 



it is desirable to keep the minimum impedance of the path about 
twice the nominal plate impedance of the output tube. At first 
thought, such a relatively low impedance path may seem unduly 
small, but in use, the effective plate impedance is considerably lowered 
by the feedback factor and further is favored by the fact that as the 
path impedance is lowered to achieve equalizations, the plate im- 
pedance is likewise lowered. Obviously, there must be a limit and 
Rll (Fig. 5) serves the purpose. Because of the foregoing facts, 
large values of gain are used, amounting to 50 db in this case, which is 
then reduced by the no equalization feedback to 30 db. The change 
in net gain is then from 24 db to 38 db, leaving 12 db minimum feed- 
back to insure stability and maintenance of characteristics. 

Theoretically, a 2-stage amplifier with only resistive feedback can- 
not oscillate even though very large values of KA are used. In 




FIG. 6. Complete system, functional schematic. 

practice, this is modified by phase shifts within the amplifier owing 
to coupling, by-pass and stray impedances so that some care must be 
exercised in design to minimize such effects. A satisfactory solution, 
when reactive elements are included in the feedback path, as in this 
application, results when the amplifier is flat in frequency character- 
istic within 2 db from 20 to 20,000 cps and no abrupt changes in 
gain exist. Nominally, the blocking condenser in the feedback path 
is made relatively large in order that appreciable changes in phase 
angle, or reduction of the feedback factor, occur at very low fre- 
quencies where the gain is decreasing and the stability requirements 
established by Black are met. The use of networks, as described, 
permits an appreciable reduction in the capacitances of the coupling 
condenser and C 3 (Fig. 5A) owing to the reverse effects of C\ which is 
made smaller than might be expected to accomplish the desired re- 
sult. This is a welcome circumstance since it reduces the total ca- 
pacitance required and thereby saves weight, bulk, and money. 



Jan., 1946 



A THREE-BAND VARIABLE EQUALIZER 



71 



Further, proper proportioning of the capacitors maintains the feed- 
back factor more nearly constant with frequency under those condi- 
tions where gain peaks are sometimes encountered in feedback am- 
plifiers. 



COMPLETE SYSTEM 



As previously mentioned, the amplifier at zero equalizations results 
in a gain of 30 db which must be reduced to zero by suitable attenu- 
ators. The system of networks is assembled as in Fig. 6 with the 
attenuators, 3 fixed and one variable (in the mid-frequency section), 



LOW FREQ 



JEfCIE! 




HIGH F 



REQUENC E:> 



FIG. 7. Low- and high-end equalization as obtained. 



split up so as to make the input and output essentially resistive in na- 
ture. 

Since the feedback network is not abnormally high in impedance 
the network controls can be placed in a separate unit which contains 
the 2 capacitor accumulator switches, the mid-frequency network at- 
tenuator with the compensating ganged attenuator, * and the feed- 
back network capacitors. Because of the potential differences which 
gather on the switch contacts, some noise will be in evidence unless 
these difference potentials are reduced by connecting resistors be- 
tween adjacent switch contacts. These bleeder resistors can be made 
sufficiently low so as to materially reduce unwanted noise without 
seriously affecting the desired frequency characteristic introduced 



72 



L. D. GRIGNON 



Vol 46, No. 1 



by each separate network on various switch steps. A further aid in 
maintaining low noise from this source is to make the attenuations 
between steps no greater than the specifications. Fortunately, the 
amount of equalization per step, the condition for low noise, and 
operating desirability are compatible. 

Fig. 7 shows 2 families of frequency characteristics as obtainable 
from the amplifier section and Fig. 8 gives the same information on 
the mid-frequency section. All combinations of these curves are ob- 
tainable. Since the mid-frequency variable network has the usual 
characteristic of varying frequency at the point of 1 / 2 attenuator loss 



Z 3 4! 5 6. 7 8,9 



MID- 



PR EQU5N 



2 3| 4| bt 6 



FIG. 8. Mid-frequency characteristics as obtained. 



the emphasis characteristic does not fully meet the specifications. 
The characteristic shown has been found satisfactory but it is planned 
to make this network of the "constant B" type, as described by Miller 
and Kimball, 1 as soon as practicable and the specification can then 
be fully realized. 

The frequency characteristics do not change with tube replace- 
ment nor with a 10 per cent change in plate supply voltage. The 
output capacity is shown in Fig. 9 for normal plate supply and for 90 
per cent normal voltage; either condition being sufficient to handle 
the output of high level reproducers or microphones. 

Total noise, measured at output of the system, is 87 db 0.001 w 



Jan., 1946 



A THREE-BAND VARIABLE EQUALIZER 



73 



which is sufficiently low to work into high gain circuits for record- 
ing or rerecording. 

The dial stop arrangement consists of 2 radial arms using the shaft 
as centers and guided on the periphery of the dial; the lock clamps 
on a machined relief also on the periphery. These arms are further 
provided with extended sections, which furnish accurate indexing, 
and pins which engage the dial pointer to provide the dial stop. 



















PER CENT DISTORTION 
- M to * U> 0> 










































"> 
o 
i 

(M 












''' 














. 






















db/.ooev 


1 


4 


/ 
2^/ + 


B 


dbM +10 +12 +14 +16 +l 



OUTPUT LEVEL 

FIG. 9. Output distortion characteristic of amplifier. 
CONCLUSION 

There has been some apprehension that feedback amplifiers with 
reactive networks in the feedback path are apt to be less stable and 
more susceptible to trouble than other conventional designs but ex- 
perience has shown this not to be the case. It is, of course, necessary 
to select suitable components to do the required job as in any other 
important work and, in this connection, it may be of interest to note 
that amplifiers designed upon the principles outlined with a frequency 
characteristic such as to replace a standard post-equalizer have been 
in service for over a year, without trouble or characteristic change, 
except for one instance caused by a faulty capacitor. 



74 L. D. GRIGNON > 

As further evidence of stability these post-equalizer amplifiers 
maintain their characteristics within V 2 db of the required char- 
acteristic, and they are not susceptible to tube replacement. There 
has also been some aversion to the use of other than constant im- 
pedance equalizers, or additional amplifiers for frequency correction 
purposes, which probably resulted from ill effects caused by changing 
impedance conditions or poor transient response; both are either re- 
duced or eliminated in the system described, as adequately indicated 
by square wave and intermodulation tests. 

The equalizer system described has met with such approval in re- 
recording work that it is proposed to use the basic design, mechani- 
cally modified to provide permanent equalizer facilities, in a new re- 
recording console to be built when conditions permit. If the mid- 
frequency section is not needed the possible use of the amplifier sec- 
tion, with the available equalizations and without the fixed attenua- 
tors, should not be overlooked as a practical method of providing a 
variable equalizer in other recording or reproducer systems where 
some gain exists or is required. 

REFERENCES 

1 MILLER, W. C., AND KIMBALL, H. R.: "A Rerecording Console, Associated 
Circuits and Constant B Equalizers," J. Soc. Mot. Pict. Eng., 43, 3 (Sept., 1944), 
p. 187. 

2 BLACK, H. S.: U. S. Patent No. 2,102,671. 

3 TERMAN, F. E.: "Radio Engineers' Handbook," McGraw-Hill (New York), 
1943, p. 395. 



PSYCHOLOGICAL AND TECHNICAL CONSIDERATIONS 

EMPLOYED IN THE BUCKY SOUND REPRODUCTION AND 

PUBLIC ADDRESS SYSTEMS* 

PETER A. BUCKY** 



Summary. Our physical senses are not capable of absolute qualitative or quanti- 
tative measurement but they indicate by comparison only. Their reaction to har- 
monious sounds is an emotional or artistic experience not measurable by ordinary 
means but definitely established as a psychological entity. Conventional theater or 
auditorium loudspeakers do not reproduce the emotional effect created by a "live" 
orchestra and it is proposed that these systems which are highly directional be replaced 
by another system with nondirectional characteristics in which reverberation from a 
multitude of separate loudspeakers replaces or even enhances the original sound 
picture. The physical difficulties involved in constructing such a system are bridged 
by the use of a radio frequency carrier for signal distribution. 

Measuring. Measuring methods are the means of comparison 
and standardization. They enable any person to duplicate certain 
findings concerning matter or energy. Psychological perception of 
matter and energy takes place by means of sense organs which, 
however, are confined to touch (consistency, shape, surface, loca- 
tion), warmth, light, sound, and smell. We have no organ of 
perception for other energies such as, for instance, certain electro- 
magnetic waves. Calibration of such energies can only be accom- 
plished by converting them into other forms of energy for which we 
are equipped with a specific sense organ. The brain does not 
perceive any physical energy directly, but transformation into 
electrochemical action is necessary which is then conducted by 
the nerves to the brain. This transformation takes place in our 
sense organs. 

Furthermore, these sense receptors are not able to differentiate 
exactly qualitative or quantitative perceptions. There are only a 
few persons who, for instance, have "absolute pitch," that means 

* Presented May 16, 1945, at the Technical Conference in Hollywood. 
** Lieut., MAC, Signal Corps Photographic Center, Long Island City, N. Y. 

75 



76 PETER A. BUCKY Vol 46, No. l 

they are able to recognize correctly the frequency of the sound 
wave to which they are subjected. We certainly can differentiate 
colors, but only by comparison. A white paper in a dark room 
illuminated by red light reflects red light only, but after some time 
we fail to perceive the color as such. We are able to differentiate 
the light intensities of various objects. We cannot differentiate 
the various components of white light, but we can generally recog- 
nize the individual frequencies of sound in an orchestra. The 
judgment of quantity is still more difficult. . At night we might be 
blinded by a match, whereas in daylight the same match does not 
interfere with our vision. 

This proves that we have no means to define by our perception 
the absolute energy and quality of light and sound. We can only 
compare qualities and quantities with relation to each other. 

Transformations of energy do not follow straight line curves, 
and these curves are by no means identical in type to the different 
methods of transformation. In physics we can coordinate the 
different curves according to some absolute measuring system. 
But when it comes to psychological perception, the problem of 
measuring sensations becomes highly complex if not hopeless. 
These considerations lead to the conclusion that our physical 
make-up gives us rather limited means of analyzing psychological 
perception. The matter becomes still more complex in the re- 
production of music as an entity in contrast to the concept of a 
heterogeneous aggregation of physical frequencies. Music as an 
art impresses the listener with emotional resonance. Besides the 
physical and physiological fundamentals, psychological reflections 
play an intricate part. It can be safely stated that the psychological 
response is the most important part of the entire problem of sound 
perception. Artistic impressions are highly individual. Therefore, 
a statistical average can be the only solution. As much as physical 
measurements might be of help, our attempts should not be domi- 
nated by them. Close cooperation of musicians, physicists, and 
psychologists is imperative, even more so than it was in the past. 

I do not pretend to assume that these deductions are novel; in 
fact, they are obvious. However, our investigations, which were 
rather empirical physically, supported these assumptions. Since the 
physical fundamentals have been so well established by the meri- 
torious work of physicists and engineers, the way is now clear for 
musicians and psychologists. To illustrate this viewpoint, a few 



Jan., 1946 BUCKY SOUND SYSTEMS 77 

particulars should be discussed from physical, psychological, and 
musical viewpoints. 

Directional Effect. The directional effect allows the listener to 
locate the source of sound waves. There cannot be any doubt that 
the artistic impression of orchestral music in general does not depend 
on the directional effect. On the contrary, nondirectional per- 
ception of sound is ideal and will produce a harmonious and mellow 
sound quality. The directional effect must be comparatively 
greater when the sound waves travel directly from the source to 
the ear of the listener without reflections or reverberations. That 
means that in an enclosed space the directional effect will be less 
noticeable. Illusion is interfered with if the directional sound 
comes from a different direction than light impressions. Therefore 
loudspeakers in motion picture theaters are usually placed near 
the screen. 

According to our experiments, nondirectional, that means evenly 
distributed sound, does not interfere with illusion. The instru- 
ments in an orchestra, especially the wind instruments, are seated 
in a linear order to dampen the directional effect and this indicates 
the fact that an overemphasized directional effect is not desirable. 
There is a great difference between the individual instruments. 
Brass instruments, for instance, are built to project the sound 
waves in a beam; whereas the majority of the other instruments 
produce unfocused spherically progressing sound waves. The 
sound of wind instruments is mainly reflected by the vertical walls, 
whereas the ceiling plays the greater part with most of the other 
instruments. If we consider the quantitative proportion of direct 
and reflected sound striking our ear, we see that there is a great 
difference between the wind instruments and the other instruments. 
However, we cannot assume that the wind instruments give a 
superior artistic effect. On the contrary, we have found that the 
suppressed directional effect of orchestra music and even distri- 
bution produces a rather pleasing effect the impression of being 
surrounded by sound and of being relieved of the effort to focus 
the attention to an individual instrument which is quite often dis- 
turbing if the directional effect has not been suppressed. 

Our experiences have led us to the conclusion that even dis- 
tribution of sound with suppression of the directional effect repre- 
sents an essential progress in the psychological artistic respect. Our 
opinion was confirmed by leading musicians and conductors who 



78 PETER A. BUCKY Vol 46, No. l 

called the effect startling. We claim that traditional assumptions 
have to be modified. We must not forget that the ideal arrange- 
ment and location of an orchestra would be in the center of a hall. 
However, the circular arrangement of the instruments would make 
it impossible for the conductor to give directions to the individual 
musicians. 

A problem in itself is represented in the reproduction of speech. 
In this case, intelligibility is affected greatly by reverberation. If 
reverberations come from comparatively long distances, the speech 
becomes unintelligible owing to time intervals of the sound waves. 
A simple solution of the problem is to use a number of low-powered 
loudspeakers distributed evenly over the listening area to reduce 
reverberation and to have the sound travel and be reflected ver- 
tically instead of horizontally in rooms or halls the height of which 
is shorter than the length. However, there exists the assumption 
that the desired illusion is interfered with if the sound does not come 
direct from the stage. This conclusion has arisen from a very 
peculiar coincidence. As mentioned above, speech and human 
voice are emitted semispherically in -a horizontal direction. When 
they are generated by the ordinary type of loudspeakers, a beam 
is emitted (as in a brass instrument) with its specific directional 
effect. If such a loudspeaker is placed in an unfamiliar location, 
no doubt the effect must be highly disturbing. We have found 
that evenly distributed sound does not appear disturbing. It is 
amazing how the listener can be misled by his imagination when 
there is no directional effect present. 

We have an installation in a room with a piano. When piano 
pieces are played over the loudspeaker system without directional 
effect, quite a number of listeners assume that the loudspeakers 
are in the piano. This proves that an action takes place in the 
brain, similar to looking at a stereoscopic picture. Psychological 
experience is substituted for physical facts. 

Another advantage is greater apparent brilliance of the sound. 
Music in an unenclosed space, such as at an open air concert, 
appears flat owing to the lack of reverberation. If the same waves 
strike the ear at imperceptible time intervals, the sound gains in 
brilliance and becomes "three-dimensional." This is a well-known 
fact. With our system, this effect is especially noticeable and 
contributes to the pleasing impression. 

Installation of System. Our system uses 3 frequency ranges 



Jan., 1946 BUCKY SOUND SYSTEMS 79 

which are amplified separately. The following rules for installation 
have been arrived at from the conclusions of our psychological 
experimentation : 

(Jf) No sound beam must be pointed directly at the listener. 

(2} The high-frequency range must be reflected from the floor by directing 
the loudspeaker toward the floor. 

(5) The medium range must be reflected from the ceiling by directing the 
loudspeakers toward the ceiling. 

(4) The low range must travel on and parallel to the floor. 

(5) In large rooms, an attempt should never be made to cover the entire space 
by means of a single loudspeaker for the individual range. 

An explanation for the resulting psychological effect, as de- 
scribed above, cannot be found by physical measurement. As we 
do not know the transmission curve of sound waves into electro- 
chemical nerve stimulation in the ear, we must restrict our observa- 
tion to empirical experiments. Up to now, we have not found a 
means for exact physical measurement. However, we can assume 
with certainty that the psychological curve does not coincide with 
the curves of the physical sound measurement apparatus. 

As our system requires a multitude of loudspeakers, a rather 
elaborate wiring system would be required. We have, therefore, 
developed a wireless system using a wavelength distant from 
presently utilized frequencies. Therefore, interferences are out of 
the question. The current coming from the microphone or the 
pickup is connected to a radio transmitter. The radio waves travel 
over the power line in the well-known way. Filters installed in the 
fuse box prevent these frequencies from reaching un desired direc- 
tions in the circuit in order not to disturb neighboring localities. 
Each loudspeaker is equipped with its own receiving amplifier and 
an adequate filtering system. This arrangement has an advantage 
in that all its parts may be manufactured in mass production. No 
special amplifiers have to be designed for the size of the hall to be 
equipped and only the number of loudspeaker units has to be 
increased. This system is, therefore, very flexible and may over- 
come comparatively easily the acoustical shortcomings of the room. 

We are confident that our system proves that for real artistic 
reproduction of sound, psychological and physiological factors are 
at least as important as physical measurements. 



TECHNICAL NEWS 



The items appearing in this section were submitted December 7, 1945, by members 
of the Technical News Committee, who welcome and will consider items of current 
technical interest from any member of the Society. 

Additional information concerning these items, or the equipment and processes dis- 
cussed, may be obtained by communicating with the general office of the Society, Hotel 
Pennsylvania, New York 1, N. Y. 

16-MM MAGAZINE FOR GUN CAMERAS 

Eastman Kodak Company. The Kodak 16-mm magazine as 
used by the Armed Forces during the recent war was designated as 
the A-6, and later the AN-A-6 magazine. Its use in the Gun Sight 
Aiming Point Camera was first for gunnery training and later for the 
purpose of recording actual combat scoring and the criticism of 
combat technique. 

The gun camera design was laid out around the standard com- 
mercial Kodak magazine, and when the camera was used for gun- 
nery training only, the magazines were loaded at the factory and 
returned to the factory for reconditioning. It soon became very ap- 
parent that pictures taken during actual combat were also invaluable 
in the training program. 

The Kodak magazine was designed for use in amateur cameras 
with spring motor drive, and for use under moderate conditions of 
temperature and vibration. It is loaded, unloaded, and repaired 
after each 50-ft run, under factory conditions by specially trained 
personnel. By this procedure the percentage of failure is kept to a 
minimum. Even when the magazine is used at high altitudes, with 
resulting low temperature and with the vibration of airplane motors 
and firing machine guns, the percentage of failures was sufficiently 
low to be considered satisfactory. 

The expansion of the gun camera program by the. Armed Forces 
was so extensive that it was impossible to train a sufficient number 
of personnel to handle the magazines properly. The Army Air 
Forces appealed to the Eastman Kodak Company to provide a new 
design of magazine which would be better suited to their facilities 
for loading and maintenance. The result was the later style AN-A-6 
magazine which has no sprocket film drive. The new design had 
80 



TECHNICAL NEWS 81 

to be suitable for use in the same cameras and yet. had to be made so 
as to be loaded and unloaded more readily and require much .less 
critical handling. 

The gun camera in normal use is subject to sufficient vibration to 
reduce the sharpness of the projected image far below that con- 
sidered satisfactory for commercial use. This fact permitted a 
design of magazine with a lower standard of steadiness than the 
commercial magazine. The design finally used is one which is 
well above the level imposed by the military operating conditions 
but will not pass the photographic tests required of the sprocket 
type. 

The sprocketless AN-A-6 magazine utilizes the camera spline 
drive solely to actuate the take-up spool. The film is moved from 
the supply spool and through the gate by the action of the camera 
film claw. A spring element is introduced between the supply spool 
and the gate for the purpose of partially leveling-off the shock of the 
intermittent action of the claw against the inertia of the roll of supply 
film. Evidently it is this resistance of the supply roll which inter- 
feres slightly with obtaining pictures of sufficient steadiness for 
commercial use, and it is for this reason chiefly that the sprocketless 
type is recommended for gun camera use only. 

TELEVISION 

The most important factor in the progress of television at the 
present time is the Federal Communications Commission's rules for 
the art and the industry reaction thereto ; the latter to be measured 
in cold dollars and cents expended in the near future. Quantity 
production and distribution of television receivers is a necessary 
part of the telecasting structure and receiver manufacturers must see 
hope for profit before engaging in large-scale production. 

Indications are that theater television will lag broadcasting de- 
velopment. Apparatus is tight now, and since the movie theater 
is* selling its regular product very successfully there appears to be 
little point in attempting to rush television as an added attraction. 

Television Productions, Inc. (Paramount) has a building in 
construction on Mt. Wilson, destined for completion by the first of 
1946. It will house their transmitter operating at a power of 4 kw 
peak, monitor equipment, living quarters, and other items. 

A 500-megacycle sight-sound beam relay will transmit the tele- 
vision programs from existing studios on the Paramount lot in Holly- 



82 



TECHNICAL NEWS 



Vol 46, No. 1 



wood to the mountain top. The equipment is expected to be in test 
operation during January 1946. 

Klaus Landsberg, Director of Television for Paramount 's televi- 
sion station, W6XYZ, brings forth the first post-war design of tele- 
vision control equipment. The Telemobile was designed along 
radically new lines by Mr. Landsberg, and constructed by his engi- 
neering staff in their Hollywood laboratory. It combines all control 
equipment necessary for the operation of 2 television cameras, in- 












FlG.l. 



The Telemobile, a"control room on wheels,"of Paramount's 
Television Productions, ready for operation. 



eluding the synchronizing pulse generator, sweep signal generator, 
power supplies, and monitoring units. 

The Telemobile is a product, neither designed on the drawing 
board of an advertising department, nor by the research group of a 
manufacturer. Instead, it is a result of practical operating experi- 
ence, and was evolved from DuMont's "suitcase" type of equipment, 
strictly out of the needs experienced by Paramount's television group 
during program operation. While functional design was the main 
consideration, and waste space was completely eliminated, the stream- 
line* design of the Telemobile establishes a new standard for eye ap- 



Jan., 1946 TECHNICAL NEWS 83 

peal. Fifteen specific advantages of this equipment are listed by 
Mr. Landsberg. 

Earl C. Anthony, Inc., is also constructing a 2-story building on 
Mt. Wilson. In addition to housing television equipment, living 
quarters and an auxiliary Diesel power room are included. 

A major studio expansion has been announced by the Don Lee 
Broadcasting System. Joining "radio row" on Vine Street in 
Hollywood, a $1,250,000 3-story studio building will be located on the 
entire block between Fountain and Homewood Avenues on Vine. 
A floor space of 105,000 sq ft will include 14 studios, offices and 
production quarters. Four theater studios with unusually large 
stages will be equipped for television as well as for radio broadcasting. 
A 150-ft tower, with elevator, will be the highest structure in Holly- 
wood. The tower will house television and FM beam relay an- 
tennae for transmission of these programs to Mt. Wilson. Here the 
company has a tract of 160 acres upon which high power transmitting 
equipment will be located when the same is manufactured. In the 
meantime, operations continue on Mt. Lee overlooking Hollywood. 
Ground will be broken in January 1946. 

PROGRESS IN THE 16-MM PROFESSIONAL FIELD 

The Technical News Committee has given a great deal of considera- 
tion to national and international activities on the production, dis- 
tribution, and use of 16-mm film during the immediate post-war 
period. 

News stories have been released which indicate that circuits of 
16-mm theaters are being planned for small towns in the United 
States, and that much of the distribution of film for the sparsely 
populated areas of foreign countries will be in 16-mm. 

The policy of the major producers of entertainment film with re- 
spect to releasing current features on 16-mm is not clear to the Com- 
mittee at this time. However, it is known that a number of inde- 
pendent producers are at present photographing on 16-mm film, par- 
ticularly in color, for release on 16-mm film to the entertainment 
field. This would indicate that 16-mm entertainment film will be 
available wherever the market exists. 

The activities of government departments in producing outstand- 
ing pictures for training and documentary purposes have firmly es- 
tablished 16-mm film in both the educational and industrial fields. 

In the educational field there is evidence to support the opinion 



84 TECHNICAL NEWS 

that the availability of strictly training and historical subjects in 
16-mm will revolutionize the teaching programs. 

The successful use of animation in training and industrial films 
has created a great deal of interest among cartoon producers who 
may find a larger market in the educational and industrial fields 
than they now enjoy in the entertainment field. 

In the equipment field much is being done toward the design and 
manufacture of professional 16-mm production equipment. Major 
improvements are being made in 16-mm color, sound, and projec- 
tion. 

An organization of 16-mm professional cinematographers has been 
formed. 

While complete details of these activities are outside the scope of 
this news item, it is indicated from the information at hand that the 
subject is of international interest. It is the recommendation of the 
Technical News Committee that a symposium on professional 16-mm 
activities be planned at an early date. 



SOCIETY ANNOUNCEMENTS 



ATLANTIC COAST SECTION MEETING 

An interesting and informative talk was given before members and guests of 
the Atlantic Coast Section of the Society by E. A. Bertram of DeLuxe Labora- 
tories, New York, on "A Motion Picture Processing Laboratory and Its Relation 
to the Film Producer." Mr. Bertram described the organization and operation 
of a modern film laboratory, covering all phases of laboratory work from mixing, 
chemicals to shipping the film. The talk was almost equal to a tour through a 
typical film laboratory. Mr. Bertram stressed the accurate sensitometric control 
and uniform development now practiced in commercial film processing labora- 
tories. 

A large audience also viewed a motion picture supplied through the Signal 
Corps Photographic Center. The meeting was held in the Penn Top of the Hotel 
Pennsylvania,. New York. 

The Chairman of the Section, C. R. Keith, thanked officers and managers of 
the Section for their cooperation in arranging meetings during his term of office. 
Mr. Keith introduced the officers and managers of the Section for terms begin- 
ning 1946, as follows : 

Chairman: FRANK E. CAHILL, JR. 

Secretary-Treasurer: JAMES FRANK, JR. 

Managers: HERBERT BARNETT JACK A. NORLING 

HOLLIS D. BRADBURY W. H. OFFENHAUSER, JR. 

G. T. LORANCE H. E. WHITE 

MEMBERS LOST SERVING THEIR COUNTRY 

The Society desires to compile a list of members who gave their lives while 
serving with the Armed Forces of their country. Such a list will include members 
abroad who served with Allied military forces as well as those in the services of 
the United States. 

The general office of the Society is not always advised of deceased members,, 
and it will be appreciated if readers of the JOURNAL will forward the name of any 
member known to them to have been a war casualty. Please include with names 
submitted the approximate date, place, and any other information available. 

Your cooperation will assist the general office in obtaining a complete and 
accurate list for the records of the Society. 



We are grieved to announce the death of Lieut. Colonel Harry B.. 
Cuthbertson, Active member of the Society, on December 12, 1945, in 
Paris, France. 

85 



86 SOCIETY ANNOUNCEMENTS Vol 46, No. 1 

EMPLOYMENT SERVICE 

POSITIONS OPEN 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 



Design engineer, experienced in mechanics and optics of motion picture 
cameras, projectors, and film scanning. Give details. Reply to Mr. 
John H. Martin, Columbia Broadcasting System, Inc., 485 Madison 
Ave., New York 22, N.Y. 

Position available for Optical Designer, capable of handling the calcula- 
tion and correction of aberrations in photographic and projection lens 
systems. Junior designers or engineers will be considered. Write 
fully giving education, experience, and other qualifications to Director 
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi- 
cago 45, 111. 

POSITIONS WANTED 

Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K,, 
354 Ninth Ave., New York 1, N.Y. 



Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 



Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 



Chief Engineer of motion picture camera 'manufacturer now available. 
Special training in optics, electricity, electronics, mechanics. Experienced 
in all phases of manufacture of cameras, projectors, and accessories. 
Prefer West Coast, but not essential. Write Robert Winkler, 119 West 
78th St., New York, N. Y. 



Newsreel cameraman, overseas U. S. Army veteran with honorable 
discharge, desires position with educational or commercial organization 
with work-training arrangement. Will supply prints on Army work in 
ETO. For full references, experience, and record, write Tom J. Ma- 
loney, 406 Oak St., Ishpeming, Mich. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 FEBRUARY, 1946 No. 2 

CONTENTS 

PAGE 

A Wide Angle 35-Mm High-Speed Motion Picture 

Camera J. H. WADDELL 87 

The Filing and Cataloguing of Motion Picture Film 

C. M. EFFINGER 103 

A Note on Chemical Drag Observed with Variable- 
Density Sound Tracks E. MESCHTER 111 

Wave Propagation and Outdoor Field Tests of A Loud- 
speaker System F. L. HOPPER AND R. C. MOODY 115 

A Film Noise Spotter J. P. CORCORAN 124 

An Integrating Meter for Measurement of Fluctuating 

Voltages H. E. HAYNES 128 

' A Survey of Photo template Methods F. DAVIS 134 

Current Literature 157 

Fifty-Ninth Semi -Annual Technical Conference 160 

Society Announcements 164 



Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office oi the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCItTY of MOTION PICTURE ENGINEERS 

MOTCL PENNSYLVANIA NW YOPVK I, N-Y- T*l. PSNN. 6 O62O 

HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 

"President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
*Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
"Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
** Engineering V ice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
* Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
""Financial Vice-President: M. R. BOYER, 

350 Fifth Ave., New York 1. 
"Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
"Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
"Treasurer: EARL I. SPONABLE, 

460 West 54th St., New York 19. 

Governors 

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
**WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairman, Atlantic Coast Section. 
**Term expires December 31, 1947. ^Chairman, Pacific Coast Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 FEBRUARY, 1946 No. 2 



A WIDE ANGLE 35-MM HIGH-SPEED 
MOTION PICTURE CAMERA* 



JOHN H. WADDELL** 

Summary. High-speed motion picture photography has played a very im- 
portant part in the development of machines and working devices by providing a con- 
venient means for analyzing the behavior of moving parts. A wide angle high-speed 
motion picture camera has recently been designed. The camera takes pictures at rates 
up to 3500 per sec and has a 40- degree field of view. This field of view allows the camera 
to photograph a 71-ftfield when 100ft away. The problems of design and applica- 
tions of the camera are discussed. 

High-speed motion picture photography has played a very im- 
portant part in the development of machines and working devices 
by providing a convenient means for analyzing the behavior of mov- 
ing parts. A wide angle high-speed motion picture camera has re- 
cently been designed to add to the photographic facilities previously 
available. The camera takes pictures at rates up to 3500 per sec 
and has a rectangular field of view 40 degrees in width. At a dis- 
tance of 100 ft this gives a field 71 ft wide. 

There have been described in papers before the Society of Motion 
Picture Engineers the 4000-picture per sec 16-mm Western Electric 
Company Fastax camera, 1 and the 8000-picture per sec 8-mm West- 
ern Electric Company Fastax camera. 2 

These qameras, as well as others available on the market, have a 
limited field of view owing to the long focal length of the lenses in 
relation to the width of the film. In the 16-mm Fastax camera, for 
example, the shortest focal length lens that can be used is 35 mm; 
and in order to show the action of a device moving over a considerable 
distance in space, it would be necessary to use a number of such 
cameras at a given magnification located so that their fields of view 

* Presented Oct. 15, 1945,,at the Technical Conference in New York. 
** Bell Telephone Laboratories, Inc., New York. 

87 



88 



J. H. WADDELL 



Vol 46, No. 2 



would overlap. This, of course, would be inconvenient and laborious. 
The new camera herein described was designed to use a 35-mm lens, 
but the film is 35 mm wide so that the field of view is much wider than 
with other high-speed cameras. Since, however, it is generally un- 
necessary to cover a wide angle of view in both vertical and hori- 
zontal directions, and because of certain design considerations, the 
pictures provided by this camera are reduced in dimension parallel 
to the length of the film as judged by conventional 35-mm film pic- 
tures. It, therefore, covers a field 12 degrees by 40 degrees. Fig. 1 



240 




80 120 160 200 240 280 320 360 400 

DISTANCE FROM SUBJECT TO CAMERA IN FEET 

FIG. 1. Field of view of high-speed cameras. 



shows a comparison of the fields of view of a number of motion pic- 
ture cameras. Fig. 2 shows the relative photographic effect for the 
16-mm Fastax and for the 35-mm wide angle Fastax; A, for equal 
distances from the subject and B for equal field widths. In A it will 
be noted that the subject sizes in the photograph are the same but 
the field of the 35-mm camera is 2 l / 2 times as wide as for the 16 mm. 
In B note that the subject size for the 35-mm wide angle camera is 
2*/2 times as large for the 35-mm camera as for the 16-mm camera. 

In previous descriptions of the Western Electric Company Fastax 
cameras, it has been noted that the shutter mechanism consists of a 
rotating prism designed to synchronize the image with the con- 
tinuously moving film. ' In the 35-mm wide angle camera the same 



Feb., 1946 



HIGH-SPEED MOTION PICTURE CAMERA 



principle is used. The wider field in this case, however, introduced 
certain design problems. The major one involves the use of a piece 
of glass about the same cross-sectional size as in the 16-mm camera 




A. COMPARISON OF CAMERA FIELDS 



16-MM FASTAX 



GUN 



SHELL 



> 



GUN 



''- 35-MM' WIDE-ANGLE CAMERA 



SHELL 



B. COMPARISON OF CAMERAS COVERING SAME HORIZONTAL FIELD 
FIG. 2. Relative photographic effects of 16-mm and 35-mm Fastax cameras. 



but of such optical characteristics as to increase the speed of the 
image by 25 per cent in order to fulfill the synchronous requirements. 
In the rotating prism type of shutter the glass must have a number 
of pairs of parallel faces and the thickness depends upon the index of 
refraction, the picture frame height, the angle of rotation in which 



90 



J. H. WADDELL 



Vol 46, No. 2 



the picture is laid down, and the peripheral speed of the image which, 
as previously indicated, must be equal to the linear speed of the film. 
For any given set of conditions this imposes a certain minimum de- 




FIG. 3. Schematic showing effect of prism rotation on ray of transmitted 

light. 

sign dimension between the rear element of the lens and the film 
plane. 

For example, in the 16-mm Fastax camera using glass with an index 
of refraction of 1.5156, the thickness of the prism is 0.551 in. This 




FIG. 4. Prism shaft assembly with ball bearings. 

is based on a frame height of 0.300 in. If this kind of glass were used 
in a full frame 35-mm camera, the prism required for the same angle 
of rotation as in the 16-mm Fastax would be 1.6 in. thick. This 
would mean a lens of much greater back focal length in a full frame 
35-mm camera than in the 16-mm Fastax, and the field of view would 



Feb., 1946 



HIGH-SPEED MOTION PICTURE CAMERA 



91 



be correspondingly reduced from that of the objective sought. 
The quantitative design of the prism, as illustrated in Fig. 3, is 
based on the following relationships : 



sin 7 
sin R 



(1} 



sm D 



b + R = i 



(2} 
(3) 



where 

SS' is the maximum displacement of ray when passing through the prism, 
/ is the angle of incidence at the prism face, 
R is the angle of refraction in the prism, 
D is the angle of displacement of the transmitted light, 
n is the index of refraction of the prism material, 
T is the prism thickness, and 
AS is the trace of a ray of light through the prism. 

In Table 1 pertinent data are given on the prisms used in the 16- 
mm and the 35-mm cameras. It is to be noted that by using a glass 
with a somewhat higher index of refraction a prism about the same 
thickness as that used in the 16-mm camera is obtained which allows 
the use of a relatively short back focal length lens in the new wide 
angle camera. The glass used is one of the rare element, high index of 
refraction, low dispersion glasses. 



TABLE 1 

Diameter of Film Pitch Circle 

Arc Film Length on Pitch Circle Correspond- 

ing to One Frame 
Angle in Degrees per Picture Frame on 

Sprocket 
Length of Chord on Pitch Circle Correspond- 

ing to One Frame 
SS' 



16-Mm Camera 
1 . 910 in. 

. 3000 in. 
18 deg 

0.2987 in. 
0.03735 in. 
1.5156 
11. 25 deg 
7 deg 24 min 
3 deg 51 min 
0.551 in. 



35-Mm Camera 
1 . 9074 in. 

. 3740 in. 
22.5deg 

0.3716 in. 
0.04645 in. 
1.7172 
11. 25 deg 
6 deg 31 min 
4 deg 44 min 
0.559 in. 



Another design problem in connection with the prism has to do 
with the mechanical support. So far only the cross-sectional size 



92 J. H. WADDELL Vol 46, No. 2 

of the prism has been considered. In the case of the 16-mm camera 
the prism is about 0.6 in. long and is supported in a metallic housing 
mounted on the end of a shaft having 2 bearings. In the 35-mm 
camera the prism is approximately 1.2 in. long and it was considered 
necessary to provide an additional bearing beyond the prism to 
eliminate whipping. Also, the increased length of the prism imposes 
increased tendencies for rupturing of the housing by the centrifugal 
forces in the housing itself. The latter problem was successfully 




FIG. 5. Interior view of camera showing rotating prism 
assembly in relation to sprocket. 

overcome by more sturdy construction. Fig. 4 shows the prism and 
its supporting system, and Fig. 5 shows the prism assembly in the 
camera. 

The sprocket design in the 35-mm camera is similar to that in the 
16-mm camera where the pitch and rectangular shape of tooth are 
designed to use film of full pitch between perforations, or film that is 
only slightly shrunken (not more than 0.5 per cent). In order that 
the user can test the lm, a gauge is provided on the back of the 
camera door. The gauge is of the "go" or "no-go" type. It consists 
of 2 pins for engaging the film perforations and which are located 38 
perforations apart. Film having a tolerable shrinkage wiH fit over 



Feb., 1946 



HIGH-SPEED MOTION PICTURE CAMERA 



93 



the pins satisfactorily. This gauge is provided as a convenience, 
because the user may neglect to refrigerate or otherwise properly store 
film that is kept on hand for a considerable time. Fig. 6 shows what 
happens when excessively shrunken film is used in a camera. The 
pictures in this figure were part of a study made on a 35-mm camera 




FIG. 



Jumping oi nlm otl sprocket at high speed 
caused by excessive shrinkage of film. 



running at 3500 exposures per sec and using film having slightly more 
than one-half of one per cent shrinkage. In this study erratic skip- 
ping of frames occurred. The high-speed motion picture studies in 
this 'figure were taken with the 16-mm Fastax running at 4000 frames 
per sec. 

Only acetate base film is recommended for use in the 35-mm camera 
and arrangements have been made with a supplier to provide it in 



94 



J. H. WADDELL 



Vol 46, No. 2 



suitable lengths packed in hermetically sealed containers. This 
together with the use of the gauge described above insures against 
trouble caused by shrinkage. 

Two view-finders are provided for use on the camera. One is a 
peep sight view-finder for use with the 35-mm lens when focused at 
infinity, and is intended as a convenient means for rapidly checking 
the field of view. The other finder is of the reflex type and fits into 
the bayonet socket mount on the front of the camera. When so used 
the 35-mm objective lens is removed from the camera and is inserted 
in the finder socket. This lens is provided with 2 focusing scales. 



4OOO 
3500 
3000 

Q 

Z 2500 
O 

o 

^2000 

K 



ft 1500 
a. 

I 

OL 

z 1000 

900 
j|j 800 
^ 700 

600 

500 
2 










































^~~^. 


~~^Z- 


ssfff 


=-=r= 














^ t 


**Z~~ 


















> 


^ 

s' 


<r* 


















A 


'/ 


'AC 


















1 


/ 
/ 
/ 
/ 





















/ 


/ 






















/ 


1 
1 






















/ 
























// 
























// 






















3O 40 50 60 70 80 90 100 110 120 130 M 



VOLTAGE ON CAMERA 

FIG. 7. Speed in pictures per second versus voltage on motors in 35-mm 
Fastax camera. 



The red scale used with the view-finder corrects for the difference in 
distances between subject and finder screen and subject and film 
plane. This scale is accordingly to be used only when working with 
the finder. The finder is provided with a 7-power magnifier as an 
aid in focusing. When the camera setup has been made the finder is 
removed and the lens replaced in the camera bayonet socket. In this 
case the white focusing scale is used and the lens is adjusted to the 
setting previously obtained on the red scale. 

The//2 35-mm lens is a standard component of the camera. There 
are also available an f/2 2-in. lens, an //3.5 105-mm lens, an //4.5 
150-mm lens, and an //4.5 254-mm lens. All of these lenses are 
fitted with bayonet mounts and can be used with the reflex finder. 



Feb., 1946 



HIGH-SPEED MOTION PICTURE CAMERA 



95 




80 90 100 

FIG. 8. Speed versus footage of exposed film at various a-c voltages. 



4000 



2000 



1500 




10 20 30 40 50 60 70 80 90 100 

EXPOSED FILM IN FEET 

FIG. 9. Speed versus footage of exposed film at various d-c voltages. 



96 



J. H. WADDELL 



Vol 46, No. 2 



All lenses are coated. It has been found by experience that an 
effective increase in aperture of approximately 2 stops is obtained by 
using coated lenses, as compared with uncoated lenses, owing to the 
reduction of internal reflections in the lens and the resulting higher 
transmission. Also the elimination of flare in the coated lenses 
facilitates the photography of incandescent subjects. 








FIG. 10. Internal view of 35-mm Fastax camera. 

The motor equipment in the 35-mm Fastax is the same as that 
used on the other Fastax cameras; namely, two 120-v, V-hp uni- 
versal motors. One motor is coupled to the film driving sprocket 
while the second motor is coupled to the film take-up spindle. The 
dynamic characteristics of the motors under load conditions are 
somewhat different in the wide angle camera from previously pub- 
lished data for the 8- and 16-mm Fastax cameras. The voltage 
versus speed characteristic in pictures per second for the 35-mm 
camera is shown in Fig. 7 for both a-c and d-c power. The speed 
characteristics at various voltages are shown in Figs. 8 and 9. The 



Feb., 1946 HlGH-SPEED MOTION PICTURE CAMERA 97 

values shown for voltage are voltages at the motors read immediately 
before the camera was started and are somewhat greater than the 
values actually existing when the motors are operating. 

A further feature in the design of the 35-mm wide angle camera is a 
flashing argon lamp timer which can be actuated from an oscillator 
furnishing a minimum of 120 v rms. The plug for the lamp is 







FiG. 11. Thirty-five millimeter Fastax camera with re- 
flex view-finder. 

mounted on the back of the camera under the driving motor. When 
using a 60-cycle voltage source the lamp provides 120 flashes per sec 
and, therefore, the markings on the film are 0.0083 sec from the 
beginning of one mark to the beginning of the next. With a suitable 
oscillator to give 100 or 1000 flashes per sec greater convenience in 
reading the film is afforded. 

The camera has been designed to withstand reasonably rough 
usage. For instance, it can be used in studies where blast pressures 
are many times greater than those that can be tolerated by man. 



98 J. H. WADDELL Vol 46, No. 2 

It is light in weight (about 35 Ib) and can be used in any position that 
insures proper positioning of the film reels on their supporting 
spindles. Since the camera can be operated on both a-c and d-c, it 
can be conveniently used in air-borne and mobile equipment, indoors 
and out, providing suitable power is available. 




FIG. 12. High-speed pictures of rocket being launched; 3500 pictures per sec. 

Fig. 10 shows the completed camera with the door removed, and 
Fig. 11, the camera with the reflex view-finder in place. 

The new camera has proved extremely useful in ballistic studies. 
Satisfactory photography in direct sunlight at //3.5 at a speed of 
3500 pictures per sec is possible. It is, of course, necessary to exercise 
care in judging the sunlight. For example, in Central United States 
and on the Eastern Seaboard, conditions for taking pictures at 3500 



Feb., 1946 



HIGH-SPEED MOTION PICTURE CAMERA 



99 



per sec exist from approximately April to September, while in Arizona, 
California, and Florida suitable pictures are obtainable all year 
around at this speed. In fact, with an optical system equivalent to 
that of the 35-mm camera sunlit pictures have been obtained in 
Florida when the subject was 8 ft under water and when the speed 
was as high as 4000 per sec. 

Fig. 12 shows several strips of pictures of a rocket launching taken 
at the rate of 3500 per sec in sunlight. The rocket was traveling 



PICTURES P 

SECOND : 

450X 




800 1200 1600 2000 2400 2800 3200 

VELOCITY OF OBJECT IN FEET PER SECOND 



3600 4000 



FIG. 13. Displacement of moving object during exposure at various 

velocities. 

approximately 1000 ft per sec in the picture. In some cases it has 
been found helpful to be able to adjust the speed of the camera in 
relation to the speed of the subject being photographed. This can 
be readily accomplished by means of a variable transformer, when 
using a-c power, and by means of a variable resistor or variable bat- 
tery when using d-c power. Fig. 13 shows the displacement of a 
moving object during the time of one exposure when photographing 
at various camera speeds, the displacement being plotted against 
velocity of the object. Fig. 14 gives the displacement of the image 
on the film per exposure when the object, moving at various speeds, 
is photographed at 3500 pictures per sec with lenses of various focal 



100 



J. H. WADDELL 



Vol 46, No. 2 



lengths. In the latter case the camera is assumed to be directed at 
right angles to the line of motion of the object and located at a dis- 
tance of 100 ft. 



OBJECT 100 FEET FROM CAMERA 
3500 PICTURES PER SECOND 




400 



80O 1200 1600 2000 2400 2800 3200 
VELOCITY OF OBJECT IN FEET PER SECOND 



3600 4000 



FIG. 14. Displacement of image on film per exposure for various 
velocities of object. Photographed at 3500 pictures per sec with lenses 
of various focal lengths. 

In addition to the more conventional applications, the 35-mm 
camera has been used for recording high-speed oscillographic traces. 
For such purposes a camera without a prism is used and the lens 




FIG. 15. Photograph of 200-kc voltage superimposed on 5000-cycle voltage 
as shown on an oscilloscope. Film speed at 120 ft per sec. 

plate is corrected to compensate for the resulting difference in back 
focal distance. By these means complete wave traces have been 
taken of alternating potentials up to 200 kc in frequency and at film 



Feb., 1946 HlGH-SPEED MOTION PICTURE CAMERA 101 

i 

speeds up to 120 ft per sec. With such pictures in a film viewer, time 
measurements are possible to an accuracy of less than one micro- 
second. Fig. 15 shows the trace of a 200-kc voltage superimposed on 
a 5000-cycle voltage. 

As already indicated, pictures made with the 35-mm wide angle 
camera described have proved very helpful in the analysis of many 
kinds of problems. Because the pictures do not correspond to the 
conventional 35-mm picture frame height, it is not convenient to view 
them on a screen by use of a conventional 35-mm projector. In- 
stead, a modification of a .standard silent projector is used. The 16- 
tooth sprockets are replaced by 8-tooth sprockets and the aperture 
plate is modified to correspond with the reduced picture height. Pro- 
visions have been made whereby such projectors can be obtained on 
the market by purchasers of the new camera. 

As indicated above, the wide angle Fastax camera offers a variety 
of possible applications. Specific examples that may be cited in- 
clude: 

(1) The firing of a gun, 

(2) Computation of pressure time curves. 

(3) The behavior of shells in flight. 

(4) The launching of rockets. 

(5) Impact of the projectile. 

(6) The landing and take-off of planes on carriers. 

(7) The launching of planes by catapult. 

(8) The behavior of planes in flight. 

(9) Wind tunnel studies. 

(10) Flight of buzz bombs. 

(11) The dropping of aerial bombs and their effects at explosion. 

(12) The vibration of plane parts. 

(13) Propeller actions. 

(14) Tow tank studies. 

(15) The launching of torpedoes. 

(16) Studies of cavitation under water. 

(17) Studies of locomotive driving wheels in motion. 

(18) Automobiles on proving grounds. 

(19) Spring and transmission studies. 

(20) The effect of combustion in fire boxes. 

(21) The spray ejection of fuels. 

The wide angle Fastax camera is being manufactured and sold by 
the Western Electric Company, Inc., as an aid in the research and 
development program of America. 

The author wishes to acknowledge his appreciation to the Wollen- 
sak Optical Company and to the Eastman Kodak Company for 



102 J . H. WADDELL 

i 

making available materials and supplies used in this camera and for 
their fine cooperation during its development. 

REFERENCES 

1 HERRIOTT, W. : "High-Speed Motion Picture Photography Applied to De- 
sign of Telephone Apparatus," /. Soc. Mot. Pict. Eng., XXX, 1 (Jan., 1938), 
p. 30. 

2 SMITH, H. J.: "8000 Pictures per Second," /. Soc. Mot. Pict. Eng., 45, 3 
(Sept., 1945), p. 171. 

DISCUSSION 

QUESTION: What type of projector is used to view the pictures taken by this 
camera? 

MR. WADDELL: This is a standard silent projector in which the sprockets have 
been changed from the normal 16-tooth to 8-tooth sprockets and by modifying the 
aperture plate to one-half the normal height, and the pictures being projected at 
16 frames per sec. 

QUESTION: I would like to know why the checkerboard background is not as 
sharp as the projectile. 

MR. WADDELL: In the first place, the checkerboard which you saw was back 
twice as far as the barrel of the gun. Now you will notice that the barrel of the 
gun was reasonably sharp. It is granted that there is a slight amount of disper- 
sion on the edges, but for all practical purposes that is negligible. But, if you 
study the plane of the subject being studied, the thing in which we were interested 
was easily measured and observed. 

QUESTION : I am curious to know what the light streaks are in front of the pro- 
jectile. 

MR. WADDELL: That, I believe, is a stick on the end of the shell which was 
used for measuring the acceleration as the shell was coming out of the barrel. 
Now, of course, these pictures were viewed frame by frame, for measuring the 
amount of travel between frames. 

QUESTION : What type of lamp do yon use in your camera for a timer? 

MR. WADDELL : We are using an argon lamp ; the neon, however, is not active 
enough for use. 

CAPT. C. H. COLES: An interesting comment, we had one of the 16-mm 
Fasta$ cameras in the altitude chamber which we took up to the pressure of 
40,000 ft altitude. We had some of the film in the tropical packing, that is the 
metalized foil and could hardly get it out of the container, because it blew up like 
a balloon. 

MR. WADDELL: We can say one thing and that is we have used this camera 
out of doors at 10 above zero in winter time and no trouble was encountered with 
brittle film.. 

COMMENT BY A MEMBER: In the matter of packing and the film blowing up in 
tropical packing, we have yet to find out how we can pack film so that, it is im- 
pervious to moisture and yet equalize pressure, and, if anyone has any suggestions 
along that line, we would be very happy to have them. 



THE FILING AND CATALOGUING OF 
MOTION PICTURE FILM* 



CARL M. EFFINGER** 



Summary. This paper describes the improved methods of cataloguing and 
filing motion picture film. While very compact, the. film catalogue system permits 
very rapid location of a single frame of the desired shot. 

Practical methods of cataloguing and filing film are the immediate 
concern of film librarians. But they also concern the industry, as a 
whole, because of the time and money saved in production costs by 
using a stock shot instead of having to shoot the scene. 

Those of us who run the Film Library at Twentieth Century-Fox 
believe that our system is the simplest and least cumbersome used 
anywhere in the motion picture industry. The old-fashioned card 
index system had been in use for many years at the studio, but an 
entirely new system was needed to save time; one that was stream- 
lined, simplified, and visual. After a good deal of experimenting, the 
system we now use was developed, and during 1932 it was put into 
practice. The system has paid ample dividends in time saved pro- 
ducers, directors, and editors, who constantly call upon the resources 
of the library. Stock shots, process background, sound effects, screen 
tests, newsreels, short subjects, March of Time films in fact, every 
sort of material that gravitates toward a film library, responds to the 
method we use. 

Under the system formerly in use at Twentieth Century-Fox, and 
still used by most of the studios, film of loosely related backgrounds 
and subject matter was spliced together, put on reels, and stored in 
the cans. When a producer or a director asked us what we had in 
the way of background of a particular nature, we would refer to the 
old card index, get out the cans indicated, and run off film until the 
customer found what he wanted, or gave up trying. With the shots 
not being broken down and segregated as to specific type, a great deal 

* Presented May 18, 1945, at the Technical Conference in Hollywood. 
** Twentieth Century-Fox Film Corporation, Beverly Hills, Calif. 

103 



104 C. M. EFFINGER Vol 46, No. 2 

of film loosely related as to subject matter had to be run. A barge 
on the Erie Canal might be wanted. Before we got to it we might 
have to look at a rowboat on the Thames, a destroyer in the Pacific, or 
a liner docking at New York, with day and night scenes all together. 
If the usable shot was at the end of a reel, the entire reel would be run 
off before we caught up with it. This cost the studio a lot of expen- 
sive time. 




FIG. 1. The catalogues are small enough to be carried Jabout and consulted by 

anyone. 

Under the present system we, get what we want in no time at all. 
A particular kind of shot is required. Only shots that come into the 
category asked for are run, and endless time is saved. Moreover, 
by breaking the material down into smaller, more exact categories, 
and making a separate roll of each, we have been able to dispense 
with the reels, which take up so much space in the can. A can holds 
1000 ft of film on a reel. Off the reel, in rolls of varying length, the 
average is 1200 ft to the can, a saving of better than 10 per cent. 

Another advantage of the system is that it is not a "one-man" 
show. Anyone who comes into the Film Library can consult the 
catalogues for himself, without calling upon the staff, and locate what 
he wants, provided the studio has it. The whole story is at his 



Feb., 1946 FILING AND CATALOGUING FlLM 105 

fingertips, visual and in text, arranged alphabetically as to categories 
and subcategories, as exactly defined and as easy to locate as any 
word in a dictionary. (Fig. 1.) 

The speed and exactness with which available material can be 
located is particularly evident when an outside studio telephones us 
regarding some specific type of shot they happen to want. While 
the person calling holds the phone, one can go straight to the cata- 
logues and in one minute or less locate everything we have that 
appertains to the subject in which he is interested. If we have what 
he wants, we take it from the can 150-ft roll, say and that is all 
he gets. This saves a good 10 min from the old way of sending an en- 
tire reel. 

Let us take a close-up of the system at work; first of all, Short 
Subjects, on which only positive film is available to us for cataloguing 
purposes. 

We receive copies of all subjects produced in. New York released 
by this company. These are run by the 3 members of the Film Li- 
brary staff, who make synopses of the usable scenes. This material 
includes newsreels, March of Time, and short subjects, such as Magic 
Carpet, World of Today, Sports Review, Adventures of a Newsreel 
Cameraman, etc. The synopses are now broken down into single 
categories for filing, so that if someone wants to see a Jap Zero being 
shot down, he does not have to look at everything appertaining to 
Jap warfare. All he would see would be subjects containing a Jap 
Zero being shot down. These categories are catalogued and filed in 
exactly the same manner as the usable material saved from our 
regular West Coast productions, which we shall discuss next. 

Both negative and positive films of our West Coast productions 
are made available to us, and we proceed as follows : 2 members of 
the Film Library staff view the picture before the negative is cut, 
and order protective master-positives on all "one- take" scenes that 
we can use that will appear in every finished product. These master- 
positives are for the Film Library files, all cut negatives being immedi- 
ately sent to New York, where the release prints are made. After 
the negative has been cut, the cut picture is viewed by at least.3 mem- 
bers of the Film Library staff. Each of the three makes a list of the 
shots he sees which he thinks could be used again and, accordingly, 
should be catalogued. 

These lists are then turned over to the catalogue girl. She makes 
a composite of the several lists, condensing the material, eliminating 



106 C. M. EFFINGER Vol 46, No. 2 

duplicate suggestions and ideas, but at the same time preserving the 
individual point of view of each of the librarians. On the basis of 
this composite list, the librarians now go through the positive film, 
in the cutting room, saving all usable scenes those in which princi- 
pals are not recognizable. Still using the composite list, they then 
examine the negative film, again saving what is usable. 




FIG. 2. The catalogue pages contain 7 frame apertures in which the film frames 
are placed for viewing. 

The next step is to match the positive and the negative film of the 
entire picture. A 4 X 6-in. filing card is prepared. On it is written 
a synopsis of the particular footage in question. To this are added 
filing data, such as production title, the names of the director and the 
cameraman, the date the shot was photographed, the "take" number, 
and the "key continuity" number. An identifying frame of positive 
film is next clipped to the card, or a negative is taken if no positive is 
available. 

The cards and the film are now turned over to the filer, who de- 
posits the film in the can, writes the number of the can on the film 



Feb., 1946 FILING AND CATALOGUING FlLM 107 

leader, and on the card, together with a one- or 2-digit number to 
identify the roll among the other rolls in the can. So that every bit 
of space in the cans may be used to the best advantage, the rolls of 
film are stored where they will fit most economically, without regard 
to subject matter. Can 13101, for instance, contains the following 
diverse shots : 

13101-1 i/L.S. Tanker Ocean Nite Fire Neg 
-2 Ship Tanker Deck Pos 
-3 Street scene Western town Pos 

-4 Warfare American Mechanized Review Desert Pos 
-5 Sound roll of a slap Neg 

-6 Newspaper office Ext. Small town Pos 
-7 Battleship in South Pacific Firing Neg 

The .larger the number, the smaller the roll, which saves time in 
locating the roll in cans. 

Positive shots, negative shots, and sound effects all may be found 
in one can. 

The cards are then classified into their categories and subcategories. 
After the classifications have been made, the cards are turned over 
to a stenographer, who lists alphabetically the essential facts on each 
card until she has a complete record of the picture to which they 
appertain. These supplementary lists are numbered, then filed. 
Their purpose is to show at a moment's notice everything that we got 
out of the picture, including sound. 

The cards are then separated into their various categories, so that 
the information on them can readily be transferred to the catalogues. 

The catalogue pages are loose-leaf so that as similar scenes accumu- 
late more sheets may be added in order to keep like material together. 
They are then divided alphabetically into main categories. Each of 
the main categories is broken down into DA WN, DA Y, D USK, and 
NITE shots, which in turn subdivide into first, second and third sub- 
categories, also alphabetically arranged. The key leaves of the main 
categories are identified by a blue tab, such as SHIPS. Some of the 
first subcategories of SHIPS, whether DAWN, DAY, DUSK, or 
NITE, are: AUXILIARY, BARGE, COASTGUARD, EXCUR- 
SION, FERRY, FREIGHTER, HOUSEBOAT, LIFEBOAT, PAS- 
SENGER, etc. These subcategories are identified by yellow tabs. 

PASSENGER, let us say, now breaks down into secondary sub- 
categories, such as, AT ANCHOR, BELLS, BOW, BRIDGE, BUOY, 
CREW, DECK, DOCK, AT SEA, and so on. These categories in 



108 C. M. EFFINGER Vol 46, No. 2 

turn break down into third subcategories which identify PASSEN- 
GER SHIP AT ANCHOR DOCK SEA, etc., as a ONE- 



3HXP - PASS - TffO STACK LIHSf> - POCKOIG - SAT 






TWENTIETH CNTUY-FOX RLM O3RPOHATION 
STOCK FILM LIBRARY 



6-2B-6794S 68109 
wooacnoN otueroft SFTIWSER STOCK 


P 10231-1 

N 10330-6 

S sA^9 


STAf. SUED LS. - S3 Washington pulling up to Dook 
L.A. barbor - en throw lines from boat to dock RU 


| JS5S3 y 


' 
' SO-DB-58087 838 

WSi 

MOOUCTIOW ' DtMttO* XF?T28 STOCK 


p 150 .JKLffl 

N 150 ^ 
S 


Ste Day "Lurline" cloaking - Los Angeles Harbor 

!' 48-DP-42478-516 
4?.-D?-'lS517-553 
. 42-DF-48555-574 
rrE. 


P 10747-3 

N 10824-5 "^W"J 

RE a! 


MtOOUCTioN j,;n. j'OTO T^KES A VACATION B'MCS FOSTER A- 170 




3ts* Day "lAirlitte" doeking {Bj'sao} "San Pedro" 
4-;!!F-43404-4S4 Keys Re?. nu6 
4-HF-43453-436 n <n 
4-3F-43437-424 " " 
4-KF -43423-402 " 
29-DP-2S694-878 " *e. 


l 10738-S 

N 10828-3 

pioo jJP^. 

N 100 sSS^i . 


mpoucnow T'R. MOTO TAJCSS A YAGA7IOK OiweaPOK F03T3R A-170 




LS Hi angle "Lurline" docking Los Angeles Harbor 

42-DF-42575-600 
42-DK -4260 1-638 
42-DF -42639 -6S3 
4S-DF-42604-688 " 
42-By-4S689-715 


P 10750-5 
N 106S2-5 

l 13S 3^ j 
N 135 





|> ^ ^^^ 

S N 


* 


P 

S 




p 


nu 


S 


net. 


p 


WOOUCTiON PHtSCfO* 





FIG. 3. A page from a catalogue showing the category at the top of the page 
description, and a frame of film for each scene. 

FUNNEL, TWO-FUNNEL, THREE-FUNNEL, or FOUR- 
FUNNEL vessel. All these subcategories run through each of the 
4 primary divisions, DAWN,, DA Y, DUSK, NITE, into which the 



Feb., 1946 FILING AND CATALOGUING FlLM 109 

main category SHIPS, like every other main category, automatically 
falls. 

Each loose-leaf page contains 7 frame apertures. (See Fig. 2.) 
Opposite each frame is a blank space for detail to be filled in on the 
typewriter. A stenographer types in its proper place the information 
she finds on the. card. She then unclips the film frame on the card 
and slips it into the vacant aperture on the loose-leaf page. It fits 




FIG. 4. One of the 2 safes in which the 36 catalogues on 22 million feet of film 

are stored. 

like a snapshot in an album. Only similar scenes appear on any 
given page. The stenographer then returns the loose-leaf to its place 
in the catalogue. When a similar scene from another picture turns 
up it will be enclosed in the vacant aperture below, or on the next 
page, if the subject matter carries over. 

At the head of each loose-leaf page is an exact breakdown of the 
shots with which that particular page is concerned. (Fig. 3.) If the 
main category were SHIPS, the breakdown might read SHIPS 
PASSENGER ONE-STACK AT ANCHOR NEW YORK 



110 C. M. EFFINGER 

HARBOR DA Y. At a glance this tells the type of shots with which 
that particular page deals. If the summary does not embrace the 
type wanted, the eye need travel no further down the page. If it 
does, the various frames of film and adjoining typewritten descriptive 
data are studied. The data consist of such information as "Shooting 
from Pier 8 Brooklyn to New York ships in f.g. New York sky- 
line in b.g. no action." This tells whoever is consulting the cata- 
logue exactly what he is looking at, and the clipping visually gives a 
general idea of what the scene is like. 

Duplicate catalogues of sound effects are furnished the Sound 
Effects Department. Also, duplicate cards of all background ma- 
terial, with frames of film attached, are turned over to the Back- 
ground Department. The staffs of these departments may thus order 
what they want direct, without coming over to the Film Library. 

An invaluable feature of our film catalogue system is its compact- 
ness. The 36 catalogue volumes are stored in 2 medium-sized safes, 
as shown in Fig. 4. They catalogue 22,000,000 ft of film. 

Two years ago Washington complimented the studio on the effi- 
ciency of our method when the joint Chiefs of Staff asked that some- 
one be sent back to the Office of War Information and the Office of 
Stiategic Services to establish a catalogue and filing system to be 
known as The United States War Film Index Library. This was to 
be a record of all war film shot by the Armed Services and by the mo- 
tion picture studios. Three systems were developed. One of them 
was essentially the manual system in use in our own Film Library. 
The other two were the IBM Mechanical System and the Electro- 
file Semi-Mechanical System; but both applied the unique features 
of our own manually operated system. 

The test of any method of cataloguing and filing film is the speed 
with which a required shot can be located. Film that cannot be 
found when it is wanted builds up picture costs. The simplicity of 
our system and the smoothness and accuracy with which it works has 
repaid the studio the investment in time and money again and again 
during the years in which it has been in operation. 



A NOTE ON CHEMICAL DRAG OBSERVED WITH 
VARIABLE-DENSITY SOUND TRACKS * 



E. MESCETER** 

During the discussion period of the meeting of the Atlantic Coast 
Section of the Society held on December 13, 1944, in New York, 
F. G. Albin gave an interesting account of a study by which 
anomalous intermodulation results were traced to chemical drag. 
With this discussion as a background, it was possible to recognize 
the effect easily when encountered in a different manner; while 
purely qualitative the direct demonstration of an effect previously 
detected by indirect means appears to be of interest. 

In the case under consideration, the 60- and 1000-cycle signal 
from an intermodulation track on an experimental material was 
impressed on the vertical plates of a cathode-ray oscillograph. A 
linear horizontal sweep, properly synchronized, permitted direct 
observation of the wave form of this signal. The oscillograph 
pattern exhibited a striking asymmetry with respect to vertical axes 
drawn through either the maximum or minimum points of the wave. 
The asymmetry apparently results from chemical drag, and has been 
accentuated by the particular combination of emulsion charac- 
teristics and developing conditions prevailing for this particular 
sample. 

Wave forms representative of the effect have been photographed 
for record purposes. The signal was applied to the horizontal plates 
of an oscillograph equipped with a blue emitting zinc sulfide cathode- 
ray tube. A transverse image of this horizontal line was formed on 
du Pont Type 201 Sound Recording Film by means of a 75-mm //2.3 
lens, the final image being about one-third the size of the original. 
The system did not include a shutter; the film was driven con- 
tinuously at 180 ft per min. This continuous motion provided a 
linear time axis, which is horizontal in the accompanying figures, 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
** Photo Products Department, E. I. du Pont de Nemours and Company, 
Inc., Parlin, N. J. 

Ill 



112 



E. MESCHTER 



Vol 46, No. 2 



in which the original oscillograms have been retouched slightly for 
reproduction purposes. 



V 



V 



FIG. 1. Original undistorted 60- and 1000-cycle signal impressed on the re- 
cording system. 

Fig. 1 illustrates the wave form of the intermodulation signal which 
is impressed on the recording system. This is a mixture of 60- and 
1000-cycle signals, the former having 4 times the amplitude of the 
latter. The amplitude of the 1000 cycles is constant at every point 






b * 
" 










FIG. 2. (o) Distorted 60- and 1000-cycle output from an experimental 
VD negative, showing asymmetry of the 1000-cycle amplitude about the 60- 
7 cycle peak, (b) 1000-cycle signal remaining aiter 60-cycle component has 
been filtered out of the fignal o 2 (a), (c) 60-cycle intermodulation signal 
formed by rectification of and removal of the 1000-cycle component from the 
signal of 2(b). 



of the 60-cycle swing, and the intermodulation of such a signal is 
zero. 

Fig. 2, however, shows an entirely different state of affairs. This 
is the wave form of the signal from a variable-density recording on an 
experimental coating. The output from the negative itself, rather 



Feb., 1946 



CHEMICAL DRAG WITH SOUND TRACKS 



113 



than from a print, has been taken to avoid possible complications 
from the printing operations in demonstrating the effect under con- 
sideration. Intermodulation values read from this film are meaning- 
less so far as practical results are concerned, but the asymmetry 
mentioned above is evident immediately. The 1000-cycle amplitude 
at B is much greater than that at C, whereas these would be equal in 
the absence of directional effects. Reaction products of develop- 
ment from the higher densities at A have streamed back and retarded 
development in the region of C, lowering the local gamma and de- 




A. 



S**** 1 



V 




FIG. 3. Distorted output and intermodulation signals similar to Fig. 2, show- 
ing even more pronounced drag effect. 



creasing the 1000-cycle response. On the other hand, B follows a 
low density region and is comparatively far removed from the pre- 
ceding high density; more vigorous development takes place here 
and the 1000-cycle response is increased. 

The oscillogram in the center is that of the modulated 1000-cycle 
signal which remains after the 60-cycle component has been filtered 
out; close examination discloses some evidence of the effect here. 
The 60-cycle intermodulation signal shown at the bottom, obtained 
after rectification and removal of the 1000-cycle component, also 
shows some horizontal asymmetry about the wave peaks. 



114 E. MESCHTER 

A still more striking example of the effect is shown in Fig. 3, where 
the asymmetry about the line AA is plainly reflected in the wave 
shape of the final 60-cycle intermodulation signal at the bottom. 
These exaggerated results have, of course, been deliberately selected 
from experimental material and should not be regarded a.s typical 
of normal conditions. 

The possibilities of further and possibly quantitative analyses of 
such oscillograms have not as yet been fully explored. However, it 
is hoped that this qualitative presentation will prove of interest and 
perhaps suggestive of other lines of approach. 

REFERENCE 

ALBIN, F..G.: "Intermodulation Distortion of Low Frequencies in Sound 
Film Recording," /. Soc. Mot. Pict. Eng., 46, 1 (Jan., 1946), p. 4. 



WAVE PROPAGATION AND OUTDOOR FIELD TESTS 
OF A LOUDSPEAKER SYSTEM* 

F. L. HOPPER AND R. C. MOODY** 



Summary. This paper discusses data and observations made with a loudspeaker 
system outdoors with regard to sound transmission as affected by such factors as wind, 
temperature, and humidity. Other data are given on measured and computed sound 
pressure distribution characteristics of multiple element radiators. 

While loudspeaker performance measurements are usually con- 
ducted indoors in rooms 1 especially designed to be as nearly totally 
absorbing as possible, outdoor measurements lend themselves par- 
ticularly to larger systems and to obtaining information as to sound 
field distribution and the intensity at some distance from the sound 
source. When these distances become great compared with the physi- 
cal dimensions of the source, the sound transmission may be materi- 
ally affected by atmospheric conditions, notably wind, temperature, 
and humidity, and by the nature of the terrain over which the sound 
propagation takes place. 

In high-powered sound systems it is common' practice to supply a 
group or array of loudspeakers instead of a single unit because of the 
limited power handling capacity of single commercial units. This re- 
quires that consideration be given to the pressure frequency distribu- 
tion since it will vary, owing to the physical separation and plurality 
of the sound sources in the array. 

This paper will present data and observations made during recent 
outdoor measurements of a large multiple source loudspeaker system 
with regard to the above mentioned factors. 

The site chosen for test was a large flat, cleared section of land in 
the Mojave desert. The loudspeaker system could be placed such 
that at least a half-mile of clear ground could be had in any direction. 
This was useful when wind was encountered, since by orienting the 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** Electrical Research Products Division, Western Electric Company, Holly- 
wood. 

115 



116 



F. L. HOPPER AND R. C. MOODY 



Vol 46, No. 2 



sound source, the effect of the wind upon transmission could be 
studied. Beyond the clearing was typical desert sagebrush growth. 
This area was used primarily for listening tests using program ma- 
terial. 

Measuring equipment consisted of an oscillator with motor-driven 
sweep and warble, a calibrated microphone, 2 amplifiers, and a vol- 
ume indicator having a logarithmic response characteristic. The out- 
put current of the latter operated a d-c recording type of milliamme- 




FIG. 1. Horizontal field for 200 cps; calculated at infinity; meas- 
ured 200 ft. 



ter. Continuous measurements could then be made at a point of fre- 
quency response, or intensity measurements for a single frequency as 
a function of azimuth or distance. A sound level meter was also used 
for the determination of absolute sound field intensities. 

The variation of sound pressure with azimuth as a function of horn 
or radiator dimensions has been adequately covered in other publica- 
tions. 3 That there is an additional variation of pressure with respect 
to azimuth when multiple sound sources are used is not as well known. 
It will be shown that when the effects of multiple sound sources are 
calculated a reasonable agreement with experimental data can be 
expected. 



Feb. , 1946 PROPAGATION AND TESTS OP LOUDSPEAKER SYSTEM 117 



Considerable work has been done in the calculation of directional 
characteristics of radiation from antennas, as well as from acoustical 
radiating systems. 4 In addition, these types of phenomena are quite 
comparable to diffraction problems occurring in optics. 5 As a basis 
for computation it is assumed that the point at which the distribution 
is to be determined is sufficiently remote from the source, so that lines 
joining this point with the elements of the loudspeaker array are es- 
sentially parallel. It is also assumed that all loudspeakers in the 
array are identical, and that the voice coil currents are equal and in 
phase. For such a system, 

the relative intensity resulting bi 

from the multiple sound radia- tij 

tors is given as |~~ ~^\ \^^-\ ' |0 

_ sin nZ 
n sin Z 



where the symbols are defined 
in the appendix. Calcula- 
tions on a large array were 
made for a frequency of 200 
cycles per sec. These data are 
shown on Fig. 1 in solid lines, 
the dotted lines showing data 
measured at 200 ft. It can be 
seen that the agreement is rea- 
sonably good in view of the 
difficulty involved in acousti- 
cal measurements of this kind. 




30 20 

ARBITRARY 



10 

DECIBELS 



FIG. 2. 



Vertical field for 200 cps (cal- 
culated). 



It should be noted, also with reference to Fig. 1, that the calculated 
data are symmetrical about the line of propagation while the meas- 
ured data show some dissymmetry. 

When an array of n columns of speakers of m speakers per column 
is to be calculated as an approximation, we may disregard all except 
one horizontal row of speakers to obtain the pressure distribution 
with regard to the azimuth, while to obtain the pressure distribution 
with respect to the vertical or zenith angle the horizontal rows may 
be neglected and only one vertical column considered. Fig. 2 shows 
the calculated vertical distribution of an array 3 speakers high. No 
measurements were made to check this distribution. 

Ordinarily the loudspeakers of an array are nearly identical in the 



118 



F. L. HOPPER AND R. C. MOODY 



Vol 46, No. 2 



frequency range covered and are driven by a single amplifier. When 
more than one amplifier is used it is possible that all speakers will not 
obtain the same voice coil current and this will modify the distribu- 
tion of an array. It is also presumed that all speakers are in phase. 
If separate amplifiers are used, care should be taken that the phase 
delay in all amplifiers is very nearly the same, as would probably be 
the case for identical amplifiers. Suppose that at some frequency one 
amplifier had a delay of one radian with respect to another amplifier. 
This time, phase difference is additive to the geometric delay and 

















































X 


x 














X 
















X 


















\ 


N 














^ 


^ 


, 












X 


v 








100 


CYCLE: 


i 


X 


s 


800 


CYC 


LE 


S 




^s 


s 


4 


DOO < 


:YC 


_EJ 






9 *S 


v 












































1 


















































X 


-. 














X 


^ 














X, 


x^ 
















X 


X 














> 


^ 















s 


s 








20 


D CYCLES 


S 


v 


2000 CYCLES 





^ . 


> 


8000 C 


YCL 


ES 






S v 
































































































X 


X? 














LOUDSPEAKER ARRAY 
ATTENUATION VS. DISTANCE 
FROM 100 TO 450 FEET 
SOLID LINES SHOW INVERSE DISTANCE SQUARE 






V 


V 


L 










D 


400 CYCLES 

1 | 


X, 


N 

































500 



100 



500 



FEET 

FIG. 3. 



Eq (1) no longer holds. In most cases the effect is to cause a dissym- 
metry about the axis of propagation. 

It is more likely that difficulty will be experienced in having an in- 
phase or out-of -phase condition as would be the case of reversed voice 
coil or amplifier connections. When the array is composed of many 
speakers it is not easy to determine the proper in-phase condition by 
listening tests. A method used in these tests may be of interest. 
Any reference speaker is chosen and arbitrarily assigned as having 
the correct phase. All other speakers are then compared one at a 
time with the reference speaker. This was done by setting up a micro- 
phone equally distant from the reference speaker and the speaker to 
be checked for phasing. A warble tone is then applied to the refer- 



Feb., 1946 PROPAGATION AND TESTS OF LOUDSPEAKER SYSTEM 1 19 

ence speaker and a volume level obtained from the microphone ampli- 
fier output. The speaker to be checked is then connected to the same 
warbled source and if the speakers are in phase the volume level will 
increase 6 db, whereas if they are not in phase, considerable cancella- 
tion results. If the warble tone chosen is the lowest frequency that 
the speakers are capable of reproducing, the difficulty of placing the 
microphone exactly the same distance from each speaker is not at all 
acute. Alternately, the microphone can be placed at some consider- 
able distance away so that the path lengths can be neglected, but the 
former method is desirable because it is more rigorous. 



E 

5 -25 

y -30 

I -35 

uj -40 

^ -45 
5 


X 








SOUND WAVE PROPAGATION 
OVER A PLANE 

SHOWING MEASURED POINTS OVER A 
FREQUENCY RANGE OF 100 TO 800O 
CYCLES AND A DISTANCE OF 100 TO 45OFT 






x^ 








^N, 










I V V 





















V<^ 


. 



















<^ 










> -55 










* ^X 


f* 



















^XQfcf 


" ^ 






3 ~ 65 

-70 












'"I 


^ s o^!? k 
>^/ 


fe 
















;, 


^^ 




d -75 

a -so, 














! .. 


^ 


&h 



















\ 


3 2O 50 100 200 500 1000 2000 5000 10000 
\ WAVELENGTHS 
FIG. 4. 



Classical theory states that the intensity of a sound wave in free 
space decreases inversely as the square of the distance from the 
source. Many observers have noted that this rule of propagation is 
altered by such factors as terrain, e. g., flat level ground as opposed to 
undulating ground or ground covered with shrubs or trees, and to 
variations in atmospheric conditions such as temperature, humidity, 
and wind. In general, these conditions become more severe as the 
distance from the sound source increases, and also as the frequency of 
the emitting sound source increases. 

Fig. 3 shows attenuation versus distance measurements made in a 
sound field produced by a loudspeaker array whose center was 7 1 /z ft 



120 F. L. HOPPER AND R. C. MOODY Vol 46, No. 2 

above flat, hard-packed clay earth. The data are given for various 
frequencies and for varying distances from the source. Measurements 
at distances less than 100 ft from the source were not taken, since it 
was desirable that the measuring distances be large compared to the 
physical dimensions of the source. For convenience of measurement 
the microphone was located about 3 ft above the surface of the earth. 
A warbled frequency was used to minimize standing wave or inter- 
ference phenomena. After calibrating the instruments, the micro- 
phone was continuously moved at a constant speed from 100 ft to 
450 ft from the sound source. During this operation a level recorder 
was used to produce a continuous record of the sound field pressure. 
The pressure decline was very constant at low frequencies but became 
more irregular at higher frequencies. The points shown on Fig. 3 are 
measured, and the solid lines are the theoretical decrease in intensity 
following an inverse distance squared law. 

The attenuation of 100 and 200 cycles shows good agreement with 
the inverse distance squared law. The 400- and 800-cycle data show 
general agreement but exhibit what appears to be a wave interference 
pattern which apparently the warble frequency did not eliminate. 
The 2000-, 4000-, and 8000-cycle data show a tendency to depart 
from the inverse distance squared law as the distance from the source 
is increased. This may result from a critical angle of incidence, or 
slight irregularities of the earth surface may have caused a dispersion 
of the reflected ray. 

If the data of Fig. 3 are replotted with wavelengths as the abscissa 
instead of feet, all data may be placed on a single curve. This is shown 
in Fig. 4. It will be observed that up to about 400 wavelengths the 
data correspond to inverse distance squared, while for greater wave- 
lengths the data show a tendency to follow an increased rate of attenu- 
ation. The curve bears a striking resemblance to the empirical curves 
of K. A. Norton 6 for attenuation of radio waves in the standard broad- 
cast band. Unfortunately, no data were obtained on low frequencies 
at distances equivalent to 500 or more wavelengths. The data are 
useful, however, in predicting sound field intensities at various dis- 
tances from the source for a loudspeaker system used outdoors, under 
similar conditions of terrain and atmospheric conditions. 

The propagation of sound may be subject to other anomalous at- 
tenuations. Knudson 7 has shown that there is an absorption of sound 
in the air caused by interaction between oxygen and water vapor. 
Knudson's data showing the absorption of sound in air with respect 



Feb., 1946 PROPAGATION AND TESTS OF LOUDSPEAKER SYSTEM 121 

to relative humidity at 20 C have been replotted in Fig. 5 with ab- 
sorption per wavelength instead of per centimeter as the ordinate. 

Under optimum conditions the attenuation of the higher audible 
frequencies even for distances as short as 200 ft may become consider- 
able. Listening tests made in the desert versus those made with the 
loudspeaker array over the ocean show a distinct improvement in 
high frequency radiation in favor of the latter condition. Other ob- 
served conditions effecting sound transmission which are undoubtedly 
not new but which furnish further confirmation of such effects, are 
briefly mentioned as follows. 



ABSORPTION VS. RELATIVE HUMIDITY 
FOR SEVERAL FREQUENCIES 




30 40 50 60 70 80 

% RELATIVE HUMIDITY AT 20 C 

FIG. 5. 



During a snowfall it was apparent that the attenuation was consid- 
erably greater although it appeared that this attenuation was not 
selective with respect to frequency. The apparent direction of sound 
seems also to change with a gusty wind, the effect being more pro- 
nounced at high frequencies. When listening to a high frequency test 
tone at some distance from the source the apparent movement of the 
source with wind direction and velocity was marked. At great dis- 
tances the sound source appeared to be as much as 20 to 30 degrees 
removed and this seems to be true for all frequencies as both voice and 
music tests were conducted. A Doppler effect was noticed, particu- 
larly when the wind direction was toward or against the direction of 
sound propagation. This produced a pronounced change of pitch 
which had a most disagreeable effect on music. An apparent variation 
of pitch as a function of pressure as noted by other observers 8 was 
distinctly heard. Musical reproduction was decidedly off key until 



122 F. L. HOPPER AND R. C. MOODY Vol 46, No. 2 

the sound field pressure was reduced to values ordinarily encountered. 
It thus appears possible with a fair degree of accuracy to predict 
the performance of a loudspeaker system as used outdoors both with 
respect to sound field intensity and distribution pattern. Other ef- 
fects of a transient nature caused by wind, temperature, and humidity 
are seen to have a marked effect upon the performance of such a sys- 
tem. 

APPENDIX 

Mathematical expressions developed by Wolff and Malter, 4 are quite useful in 
predicting the directional response of a multiple source loudspeaker array. 

For a combination of n equally spaced linearly arranged sources, having equal 
intensity, and moving in phase, the intensity at a remote point A is given as 



_. sin n Z 

Fa. = : ~ (-Z) 

n sin Z 



where 



d is the separation between individual radiators, X wavelength, and a the angle 
subtended between a line perpendicular to the linear array of loudspeakers and 
point A . 

The intensity is a maximum in a direction perpendicular to the linear array 
when all sources are in phase, since a = and sin a = and the value of F a in 
Eq (7) is unity. 

For a straight line source having uniform intensity and all points in phase Eq 
(1} becomes 

F Sin Z f9\ 

Fa = --- (5) 

since n approaches infinity and d approaches zero. 

Other more complex expressions are developed for cases in which there is pro- 
gressive phase shift along the array, and for cases where the points are in phase 
but the intensity is nonuniform. Uusally the first condition would not exist for an 
array in which identical loudspeakers are used, but might occur were the loud- 
speakers placed not in a line, but staggered with respect to each other. In the 
second case, nonuniform intensity could easily result were the loudspeakers oper- 
ated from amplifiers having unequal gains, or from any cause resulting in different 
currents flowing in the individual voice coils of the loudspeakers. 

REFERENCES 

1 BEDELL, E. H.: "Some Data on a Room Designed for Free Field Measure- 
ments," /. Acoust. Soc. Amer., VIII (Oct., 1936), p. 118. 

2 HOPPER, F. L., AND ROMANOW, F. F.: "Determination of Microphone Per- 
formance," J. Soc. Mot. Pict. Eng., XXXVI, 4 (Apr., 1941), p. 341. 



Feb., 1946 PROPAGATION AND TESTS OF LOUDSPEAKER SYSTEM 123 

3 MCLACHLAN, N. W.: "Loudspeakers Theory, Performance, Testing and 
Design," Oxford Press, 1934. 

4 WOLFF, I., AND MALTER, L.: "Directional Radiation of Sound," /. Acoust. 
Soc. Amer., H (Oct., 1930), p. 201. 

5 HOUSTOUN, R. A.: "A Treatise on Light," Longmans-Green and Co. (New 
York), 1938. 

8 NORTON, K. A.: "The Propagation of Radio Waves Over the Surface of the 
Earth, and in the Upper Atmosphere," Proc.I.R.E.,24 (Oct., 1936), p. 1367; 25 
(Sept., 1937), p. 1203. 

7 KNUDSON, V. O. : "The Absorption of Sound in Air, in Oxygen, and in Nitro- 
gen. Effects of Temperature and Humidity," /. Acoust. Soc. Amer., V (Oct., 
1933), p. 112. 

8 FLETCHER, H.: "Loudness, Pitch and the Timbre of Musical Tones and 
Their Relation to the Intensity and the Frequency and the Overtone Structure," 
/. Acoust. Soc. Amer., VI (Oct., 1934), p. 59. 



A FILM NOISE SPOTTER* 
J. P. CORCORAN** 



Summary. A machine is described by which defective splices, dirt, and "pops'' 
are detected before running the film through a regular sound reproducer. 

In the translation of the useful signal from film sound tracks to 
audible sounds, quite often other undesirable signals are scanned by 
the light beam. These undesirable signals, commonly referred to as 
"pops," are caused from several types of imperfections in the track 
area. The types frequently encountered are small clear spots on the 
print, usually resulting from foreign particles on the negative. Other 
types may be caused by imperfect splices, scratches, and recorded noise, 
such as clicks, etc. 

In the preparation of rerecording tracks these imperfections must 
be detected and painted out to prevent unnecessary delay during the 
rerecording process. The past practice has been to inspect each foot 
of the sound tracks under a magnifying glass. This method was very 
tedious, usually resulting in severe eyestrain. In addition, consider- 
able inspection time was required with no particular assurance that 
all the "pops" had been painted out. 

Upon investigation to determine whether the film cutter could be 
supplied with some tool by which this work could be performed with 
less fatigue and a saving of time, it was learned from Robert Cook of 
the Disney Studio that reasonable success had been achieved by the 
adaptation of a Moviola sound head. With this suggestion we built 
a small, compact sound reproducer unit which, being portable, could 
be placed between the rewinds on a film cutting table. This unit is 
shown in its operating position in Fig. 1. Four flanged rollers, two of 
which are located on each side of the scanning drum, are offset to re- 
tain the film in focus as it is pulled past the scanning aperture by the 
rewinds. The exciting lamp, which is located in the housing just above 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** Twentieth Century-Fox Film Corporation, Beverly Hills, Calif. 

124 



A FILM NOISE SPOTTER 



125 



the drum, is mounted on the objective lens barrel. This method per- 
mits the rotation of the lens while still retaining lamp focus. 

In order for the "pops" and other undesirable signals to be dis- 
tinguishable by the use of the earphones, it is necessary to suppress the 
normal signal. This can be accomplished in either of 2 ways, depend- 
ing upon the type of track the operator wishes to inspect. If the oper- 
ator is dealing with push-pull track, normal signal suppression is 
achieved by signal cancellation in the photocell circuit when adjust- 
ing the scanning beam azimuth to be parallel to the signal striations 




FIG. 1. Film noise spotter developed by Twentieth Century- Fox Studios. 



on the film. For standard track, sufficient reduction of the normal 
signal is realized by rotating the objective lens barrel to approximately 
25 degrees off normal scanning azimuth. The mask, which permits 
the scanning of 100-mil or 200-mil tracks, is located in the scanning 
region inside the drum. It is operated by rotation of the smooth knob 
on the end of the drum. 

A photocell is located inside the drum and, as noted in Fig. 2, the 
signal is increased to earphone level by a 3-stage amplifier. The ex- 
citing lamp as well as the amplifier is a-c operated. A filter is located 
in the cathode circuit of the first stage to reduce the 60-cycle hum be- 



126 



J. P. CORCORAN 



Vol 46, No. 2 




Feb., 1946 A FILM NOISE SPOTTER 127 

low an objectionable level. A volume control is also provided for the 
operator's convenience. 

The rate of film travel is approximately 100 ft per min. When a 
"pop" is detected, the film is stopped, and moved forward and back- 
ward by the use of the rewinds until the spot is located at the scan- 
ning light. The film is marked and then pulled forward over the in- 
spection glass and the spot is painted out. Quick drying ink, which 
dries in about 3 sec, is used. 

This film noise spotter has proved to be a popular tool with the 
film cutters whose work deals essentially with the preparation of the 
rerecording tracks . It is safely estimated that the work which required 
8 hr to do by the usual inspection method, can, with the use of this 
device, be performed in one hour. In addition, the cutter has the satis- 
faction of knowing that all the imperfections which would give trouble 
in the dubbing monitor room have been painted out. 



AN INTEGRATING METER FOR MEASUREMENT 
OF FLUCTUATING VOLTAGES* 



HAROLD E. HAYNES' 



Summary. As a supplement to other types of meters for measuring voltages of 
fluctuating amplitude, one was developed which integrates the voltage being measured 
over a chosen interval (a few seconds}, and at the end of the interval indicates the 
average value that has existed. The laboratory model of the meter is described and its 
operation explained. Some possible applications to sound engineering are suggested. 

REASONS FOR DEVELOPMENT 

Mahy of the alternating voltages which must be evaluated in con- 
nection with sound studies are quantities that fluctuate at rates easily 
perceptible to the human senses. Ground noise in film recordings, 
surface noise and turntable rumble in disk recordings, ambient noise 
in various locations, and speed variations in sound recording and re- 
producing apparatus, are examples of this type of function. The pur- 
pose of making such measurements is usually to assist in analyzing 
the defects in equipment and processes that cause them, and for that 
purpose it is desirable to break down the over-all effect into compo- 
nents whose causes may be traced individually. However, for con- 
ciseness it is frequently desirable to express such a quantity by a 
single numerical value, even though such .a value does not tell the 
whole story. This is particularly true in factory tests of equipment 
or processes where the chief objective of the measurement is to decide 
whether certain standards have been met, and more elaborate tests 
are not justified, or in comparing development models of a certain 
piece of apparatus, where it is necessary to decide which of several 
designs is preferable, and a "lump sum" evaluation may be very use- 
ful. 

The ordinary indicating meter, while often satisfactory, requires 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
** RCA Victor Division, Radio Corporation of America, Indianapolis, Ind. 

128 



METER FOR MEASUREMENT OF VOLTAGES 129 

that the observer do sorrie of the necessary averaging in his head, and 
if the amplitude of the voltage being measured is fluctuating quite 
widely at frequencies in the range of 0.5 to 10 cycles per sec, inter- 
preting the action of the meter may be troublesome and the subject 
of disagreement among observers. An integrating meter, that is, one 
which measures volt-seconds for a time interval of appropriate length 
and registers only a single value proportional to average amplitude 
during the interval, was thought desirable as an auxiliary instrument 
in making these measurements. 

Development of the voltmeter herein described was prompted by 
the need for a meter which would give accurate, consistent measure- 
ments of disk record surface noise in the short interval of time oc- 
cupied by the unmodulated groove at the beginning or end of a disk 
record, without being subject to the vagaries of the observer in inter- 
preting a fluctuating reading. When an indicating meter is used, a 
compromise must be made between heavy and light damping. Light 
damping permits the pointer to follow to some extent the amplitude 
variations of the voltage being measured (and the meter will over- 
emphasize fluctuations near its own mechanical resonance frequency), 
and damping heavy enough to produce a satisfactorily constant indi- 
cation requires a signal of longer duration than that which may be 
available in order to reach its ultimate reading. 

The fluctuating a-c voltage to be measured must first be rectified 
(as is done electrically or otherwise in every a-c voltmeter) so that a 
certain amplitude will produce a given effect irrespective of its polar- 
ity. The rectifier may produce a d-c voltage proportional to the first 
power of the input voltage, or to the square or some other power of 
it ; or it may produce a voltage proportional to the peak value of the 
a-c voltage. In order for the integrating meter to be directly compar- 
able with the standard-type volume level indicator, it was designed to 
measure average value. However, the way in which an integrating 
meter responds to instantaneous voltage values is not restricted to 
average reading by its integrating action, as explained above. It is 
perfectly possible to design an integrating meter which reads average 
squared value, or average peak value, just as easily as true average 
value. Which of these types of action the meter should have involves 
psychological considerations, since the problem is one of evaluating a 
physical quantity in terms of its effect on the senses. It certainly de- 
pends upon the kind of quantity to be measured, and is, of course, 
debatable for any one kind. 



130 H. E. HAYNES Vol 46, No. 2 

PRINCIPLES OF CIRCUIT OPERATION 

If a capacitor is charged for a given interval from a source of fluc- 
tuating voltage e = /(/) through a resistance R, the voltage across 
the capacitor at time / will be e c = ^J^idt. If i is kept proportional 
to e, e c at time / will be accurately proportional to the integration of e 
over the interval; this can be accomplished by insuring that e c is 

always small compared with e, since i = - c . In the circuit de- 

R 




FIG. 1. Integrating meter for the measurement of fluctuating voltages. 

scribed, e c reaches only V 20 the average value of e; if for some voltage 
of unusually low form factor it were found necessary to reduce this 
fraction to a lower value, such a change could easily be made. 

The meter, shown in Fig. 1, consists of a voltage amplifier, a phase 
inverter, a full- wave rectifier, and a capacitor charging circuit of the 
type described, with the necessary timing circuit, and high-impedance 
d-c voltmeter. No particular attempt was made to make the meter 
an instrument of great absolute accuracy, since it was expected that 
its chief application would be in comparing 2 voltages (such as full- 
track modulation and ground noise) with attenuators providing the 
major part of the numerical measurement. However, stability is en- 
tirely adequate and calibration simple. 



Feb., 1946 METER FOR MEASUREMENT OF VOLTAGES 131 

A valuable feature of the meter, though not peculiar to an inte- 
grating type, is its ability to withstand without damage to itself an in- 
put level far higher th#n full scale. This makes it unnecessary to take 
special precautions such as disconnecting the meter when high-level 
inputs are imminent, as, for example, when going from unmodulated 
to modulated sound track. 

Frequency response is essentially flat from 50 cycles to 15,000 cy- 
cles, so that the meter may be used for any of several types of meas- 
urements by combining it with appropriate niters. 

DESCRIPTION OF CIRCUIT 

A brief reference to the circuit diagram, shown in Fig. 2, will serve to 
clarify its operation. One section of the 6SN7GT is a voltage ampli- 
fier, and the other acts as a phase inverter so that full-wave rectifier 
action can be obtained without the use of a transformer. A poten- 
tiometer in the input circuit permits calibration. The rectifier de- 
velops a fluctuating unidirectional voltage across R it and this voltage 
charges C\ through a high resistance. (This is ^2, Rz, or R, depend- 
ing upon the setting of the time interval selector switches Si and 2, 
which are ganged so that properly related time constants will exist 
in the integrating and timing circuits at all times.) The voltage across 
C\ is indicated by a vacuum-tube d-c voltmeter that has extremely 
high-input impedance, consisting of the two 6 V6 tubes in a balanced 
circuit of exceptional stability. 

Timing is accomplished by another resistance-capacitance cir- 
cuit, the 2050 thyratron, and a relay, in the following manner : when 
the switch 63 is thrown to "start" (the upper position in the diagram), 
C 2 begins to charge from the stabilized d-c voltage appearing across 
the VR105, through R b , R 6 , or ^7 (depending on the interval desired) . 
The voltage across Cz appears at the grid of the 2050 thyratron, while 
its cathode is maintained at +60 v (before firing) by a voltage di- 
vider across the VR105. When the grid potential has risen to +58.5 
v,* the 2050 thyratron ionizes, energizing the relay and breaking the 
signal-integrating circuit. 

* Note that the time interval is determined almost entirely by R and C values, 
since the ultimate capacitor voltage is far greater than normal variations in the 
critical grid voltage for the thyratron (which is approximately 1.5 v). Both 
cathode and grid voltages are derived from the same d-c supply, causing any varia- 
tions in this supply to produce only second-order effects on timing. 



132 



H. E. HAYNES 
OPERATION 



Vol 46, No. 2 



The proper input level for full-scale deflection is approximately 
one volt rms (steady sine wave), and calibration is readily accom- 




plished if desired by means of a sine- wave input of known amplitude 
and any convenient frequency within the range of the instrument. 



Feb., 1946 METER FOR MEASUREMENT OF VOLTAGES 133 

Measurements are made by throwing the switch S 3 (normally at 
"off") to the "start" position. Timing action then takes place, as ex- 
plained above, leaving a voltage across C\ which is proportional to the 
integrated value of the charging voltage which appeared at the recti- 
fier output during the interval. Owing to the extremely low current 
required by the measuring circuit, the meter reading remains vir- 
tually unchanged for several seconds (dropping about 5 per cent in 
30 sec). This allows data to be taken in a leisurely fashion. After 
the meter has been read, the switch is returned to "off," which dis- 
charges both C\ and C 2 , thus preparing the meter for another cycle 
of operation. 

Convenient time intervals may be from 0.5 sec to 5.0 sec, depending 
upon the specific application. It is only necessary to make the tim- 
ing circuit constants (corresponding to C 2 and ^ 5 ) satisfy the relation 
58.5 = 105(1 -e ~ t/RC ), where / is the time interval desired, and to 
make the integrating circuit constants (corresponding to R 2 and Ci) 
satisfy the relation RC = 20t. In the instrument constructed in our 
laboratory, the intervals chosen were 1.80 sec, 1.54 sec, and 5.0 sec, 
the first two corresponding to one turntable revolution at 33 Vs rpm, 
and 2 revolutions at 78 rpm, respectively. For ground-noise measure- 
ment, a representative sample may be obtained in one second or less. 
This integrating meter is directly applicable to flutter-measuring 
equipment of the Wheatstone-bridge variety, which indicates speed 
variations as amplitude modulation of an audio-frequency carrier. 
For this application, a longer interval is desirable, probably several 
times the period of the least frequent speed disturbance to which the 
reproducing apparatus is subject. 

Utilization of the integrating meter has been found convenient and 
helpful in the evaluation of fluctuating alternating voltages when a 
few-second average can be of value. Since the principles of its opera- 
tion are simple and straightforward, the basic design is flexible and 
can be adapted to somewhat different requirements. 






A SURVEY OF PHOTOTEMPLATE METHODS* 
FAUREST DAVIS** 



Summary. This survey covers phototemplate methods now in use in the Southern 
California airframe industry. These methods are concerned with the photographic 
reproduction of original engineering line drawings. Although the methods outlined 
are most extensively used in the manufacture of aircraft at the present time, they 
can be readily adapted to other fields where the rapid, accurate, and economical 
reproduction of engineering information is required. Engineers in the airframe 
industry are certain phototemplate methods will be indispensable in meeting post- 
war competition; it is thought other industries with similar line reproduction problems 
will find these methods indispensable to their post-war operations. Accordingly, 
working details on the following procedures, which have been proved by heavy pro- 
duction use, are given for the information and guidance of engineers: excitation, 
camera, and reflex methods; other satisfactory but less extensively used methods; 
and practical methods for applying gelatin-silver-halide emulsions to surfaces at 
the work site. The survey concludes with a table giving comparative cost and pro- 
duction data on the 3 major methods. 

This survey covers phototemplate methods now in use in the air- 
frame industry in Southern California, where the variety of opera- 
tions has brought about development of a number of methods for 
making photographic reproductions of original engineering line 
drawings. The primary requirement of such reproductions is high 
size fidelity to the original, which must itself be dimensionally stable. 
Consequently, most originals and reproductions are made on metal, 
which so far. is the most practical material having low size-change 
characteristics. The methods described below not only meet the 
requirement for accuracy, but are more economical and rapid than 
the hand layout methods formerly used. It is hoped that the basic 
working procedures given in this paper will be useful to engineers 
with line reproduction problems. All materials and methods de- 
scribed are in production use. 

In the early days of phototemplate procedures (1940) it was planned 
to have sheet metal sensitized in the Midwest and shipped to the 

* Presented May 18, 1945, at the Technical Conference in Hollywood. 
** General Aniline and Film Corporation, Ansco Division, Los Angeles, Calif. 

134 



SURVEY OF PHOTOTEMPLATE METHODS 135 

Pacific Coast. The impracticability of this idea was soon evident, 
and fast sensitive materials and means of applying them to surfaces 
at the work site were sought. 

PHOTOSENSITIVE PRODUCTS 

Bulk Sensitizing Materials. The Hammer Dry Plate Company 
has a bulk gelatin-silver-halide emulsion, which is used for spray 
application. This emulsion is of the colorblind process type and is 
suitable for ordinary photographic contact and projection use. 

Ansco has 3 types of bulk emulsions available for industrial use. 
The first of these, Regular Type, is used for ordinary photographic 
contact and piojection purposes, and has a speed about equal to that 
of Brovira No. 2 paper. The second, Type X, was designed espe- 
cially to meet the requirements of excitation methods to be de- 
scribed. It is about 50 per cent faster than the Regular Type, is 
richer in silver, and is more contrasty. The third is Type RC, a 
negative reflex emulsion for those industries whose operations are 
large enough to warrant their setting up to sensitize prepared plastic 
or other translucent media instead of using reflex paper. This 
emulsion is the same as that used by Ansco in sensitizing its Reflex 
Paper 2505 described below. One kilogram of these emulsions 
produces slightly more than a quart when liquefied. 

Prepared Sensitized Materials. Some operations use sensitized 
stripping material in roll form for application to prepared metal 
sheets. 

The Eastman Kodak Company has a product called Linagraph 
Transfer Paper, which consists of a calendered paper base of high 
tear strength coated with a cellulose stripping layer, over which is 
coated a silver halide emulsion. By means of a special motor-driven 
laminator this material is laminated to metal sheets, which have been 
prepared with a soft nitrocellulose lacquer. The laminator carries 
the lacquered sheet over a roller soaked with solvent to make the 
lacquer surface tacky, and passes it into rollers where the transfer 
paper is applied emulsion to the lacquer. Several hours are required 
for the lacquer solvent to escape through the paper base, which may 
then be stripped from the emulsion. The sheet thus sensitized is 
ready for exposure by ordinary photographic contact and projection 
methods, or by excitation methods. 

The construction of the laminator limits the thickness of material 
that can be readily sensitized to V 4 in. In modern photographic 



136 F. DAVIS Vol 46. No. 2 

tooling methods a single order often calls for the sensitizing of several 
different types of materials with thicknesses of 0.036 in. (Alclad for 
prototype), 0.25 in. (plowsteel for dies), and 1.5 in. (Masonite or 
Richlite for formblocks). 

The tooling department of one airframe plant uses the time 
honored bichromated colloid as a sensitizer, which is applied to a 
metal sheet in a whirler in the usual manner. Further details will be 
found in the discussion of a lithographic method. Theses ensitizers 
can be made up by anyone; formulas can be found in any lithogra- 
phers' manual. 

Ansco Reflex Paper 2505 is used extensively in the reflex method. 
It has high inherent contrast, clean-working quality, and sensitivity 
to red as well as other regions of the visible spectrum. 

Defender Lithographic Paper is an orthochromatic paper satisfac- 
tory for reflex use. 



APPLYING EMULSION AT THE WORK SITE 



Spray Sensitizing. This method was developed at Lockheed 
Aircraft Corporation and is now used in many airframe plants. It 
requires ordinary paint spraying equipment with the exception that 
the inner parts of the spray gun are of stainless steel or are silver 
plated; any rubber tubing through which the emulsion passes, as 
from pressure pot to spray gun, must be free of emulsion contami- 
nants such as sulfur compounds. Operations must of course be 
carried out under proper darkroom safelight conditions, 25-w ruby 
lamps being satisfactory in most cases. 

Bulk emulsion of the gelatin-silver-halide type is shipped from 
manufacturers in refrigerated insulated crates and is stored at about 
45 F. It arrives in gel form, lumps or noodles, and must be gently 
warmed in a water jacket to 100 F to liquefy it. Glass, enamelware, 
or stainless steel vessels should be used as containers for the emul- 
sion. The liquefying temperature is not critical, 95 to 110 F; it is 
not advisable to go much beyond 110 for prolonged periods, say an 
hour or so, because of the danger of building up a high fog level in the 
emulsion. 

Metal sheets are prepared for sensitizing by washing them down 
thoroughly with wash thinner and wiping dry with clean cloths. If 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 137 

the metal is intended for tooling operations, such as shearing, punch- 
ing, bending, etc., a single coat of highly plasticized pigmented lac- 
quer is applied, allowed to dry to touch, and immediately followed by 
2 coats of a highly pigmented white or tinted matte lacquer contain- 
ing less plasticizer. As far as phototemplate operations are concerned 
there is no objectionable halation from a sensitized white lacquered 
board. These last lacquer coats must dry to handle in 5 min and 
within a few hours set up to a hard coat in order to hold the gelatin 





FIG. 1. Spray-sensitizing installation at Lockheed Aircraft Cor- 
poration. 

of the emulsion in place after it is hardened in processing and dries 
out on aging, when it exerts a tremendous pull on its lacquer support. 
The emulsion is sprayed directly onto this matte lacquer, and it is 
important that the lacquer be entirely free from sulfur compounds or 
other contaminants, otherwise serious fog troubles will be encoun- 
tered. For proper control of metal preparation, it is desirable to have 
painting operations under phototemplate department management. 

Facilities as required by most state laws for the spraying of lac- 
quers are entirely adequate for carrying off any silver halides which 
would otherwise be dispersed into the workroom atmosphere during 
spray sensitizing. A satisfactory installation consists of a spray 



138 F. DAVIS Vol 46, No* 2 

booth with proper light-tight input and exhaust fans. It has been 
found desirable to have the input volume per minute exceed the 
volume of the workroom, which should be entirely closed during 
sensitizing, by 25 per cent. In this region this input supply does not 
have to be filtered. The exhaust should carry off a volume of air per 
minute equal to the volume of the workroom. Thus a positive pres- 
sure is built up to force dispersed silver halides immediately out 
through the exhaust. Three banks of metal filters should be put into 
the path of the exhaust at the back of the spray booth to catch 
residual spray and keep it from passing into the outside atmosphere. 
The installation of a water bath in the spray booth is not necessary. 




FIG. 2. Brush-sensitizing at Douglas Aircraft Company. 

Metal sheets or other materials are placed on an A -rack at an 
angle of 15 deg to the vertical and hand-sprayed with a vertical over- 
lapping stroke. The pressure pot is used as a water jacket into 
which is placed the inert vessel containing liquefied emulsion, which 
is held at about 100 F for proper spraying viscosity. The spray gun 
is adjusted to give a flat "fan" about 8 in. wide when the gun is held 
about 8 in. from the surface. A hand flashlight with a red filter is 
taped to the spray gun to permit inspection as the sensitizing pro- 
ceeds. Air pressure on the pot is 8 Ib, just sufficient to force the 
emulsion through the line to the spray head, where a second air line 
supplies 40-lb pressure to blow the emulsion into a fine spray. (See 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 139 

Fig. 1.) A proper spraying technique results in no waste from run- 
off, but if there is any, it may be caught in a stainless steel trough 
and used over again. 

For a 4 X 12-ft sheet of metal the spraying cycle is 3 min, includ- 
ing placing the material on the rack, spraying, and moving the wet 
sheet to a drying compartment, which contains heating coils, ther- 
mostatically controlled to hold temperature in the cabinet to 95 F, 
moisture-absorbing cellulose pads, and a blower. Notice that no 




FIG. 3. Vertical tanks set in the floor are most efficient for proces- 
sing large sheets of sensitized metal. A hoist is required for handling 
heavy pieces as shown in this installation at Consolidated Vultee Air- 
craft Corporation. 



chilling is required. The sheet is ready for use or storage in 20 min. 
In this method of applying emulsions a gallon covers 250 sq ft. 

Brush Sensitizing. The Douglas Aircraft Company's plant at 
El Segundo, California, did not wish to set up a spray installation be- 
cause of long delays necessitated in obtaining equipment. Bulk 
emulsion was best suited to the great variety of their requirements, 
and a method was soon worked out for its application to lacquered 
surfaces with a soft paintbrush. (See Fig. 2.) For this method the 
emulsion is liquefied and brought to 110 F, when a Vrhp 



140 F. DAVIS Vol 46, No. 2 

Eppenbach Homo-Mixer or equivalent is placed in it and run at 5000 
rpm; 7.5 cc butylcellosolve per 100 cc liquefied emulsion are added 
and the Homo-Mixer run for 3 min. A certain amount of foaming 
will occur during homogenizing and sufficient Foamex or other inert 
defoamer should be added in drops to eliminate it. The butylcello- 
solve acts as gelatin leveling agent and the Foamex in addition to de- 
foaming will prevent air bubbles from being pulled out of the brush 
into the emulsion coating. Glycols other than butylcellosolve promise 
even better leveling, but these are at the moment difficult to obtain 
in required production quantities. An emulsion so treated can be 
resolidified and held for long periods without the finals separating 
out, and footage obtainable is truly surprising: 1000 to 1500 sq ft per 
gal! This brings the cost of sensitizing material to about one cent 
per sq ft. 

In the case of the 1500-sq ft coverage there was a slight loss of 
speed and maximum density, but these are not important to the 
phototemplate method. One operation uses a silk screen for apply- 
ing an emulsion similarly treated, and obtains similar coverage. 

PROCESSING 

Tanks. Experience has shown that the most efficient installation 
for processing large sensitized and exposed materials is a bank of 4 
vertical tanks set side by side in a well in the laboratory floor. The 
tops of the tanks are usually at or somewhat above the general floor 
level. (See Fig. 3.) Three-inch tongue-and-groove Douglas fir, 
cypress, redwood, or spruce are practical construction materials for 
this purpose. Heavy stainless steel is also used. Any plumbing is 
of stainless steel or hard rubber. Assuming a maximum sheet size 
of 5 X 12 ft, each tank has inside dimensions of approximately 5 in. 
X 6 ft X 13 ft, and has a volume of about 250 gal. In the excita- 
tion and camera methods, D-ll is most generally used, but in the re- 
flex method the following formula is often used for both negatives and 
prints : 

Formula AP-80 

Water 125 F 2 1 /* gal 

Metol 300 grains 

Sodium Sulfite Anhydrous 3 Ib, 2 oz 

Hydroquinone 7 l /% oz 

Sodium Carbonate Monohydrated 1 Ib, 14 oz 

Potassium Bromide 4 oz 

Water to make 5 gal 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 141 

These developers are kept to volume by adding fresh stock de- 
veloper and will last from 3 to 6 months before complete renewal is 
required. Development is by inspection and is between 3 and 5 min 
at 68 F. Some operators install mechanical agitators in their de- 
veloper tanks, whereas others prefer merely to stir the solution up 
from the bottom with a long pole once or twice a day. 

In some localities the range of temperatures is such that it is neces- 
sary to heat or cool the solutions. In regions where the temperature 
range is short the tanks are set in a well-insulated well with sufficient 
free air space around them to permit circulation of heated or cooled 
air automatically supplied and thermostatically controlled, at a cost 
of $200 to $500. However, where the temperature range is consider- 
able, it will be necessary to set up a rather elaborate installation to 
circulate the solutions through heat exchangers with the cost rang- 
ing between $3000 and $5000. 

To stop developer action and prolong fixer life, a solution of 2 per 
cent acetic acid is recommended as a stop bath. This bath wil^ 
gradually lose efficiency as developer is carried into it. A rough 
check can be made periodically with litmus paper to determine 
whether the bath has lost its acidity. When it has, it can be entirely 
replaced cheaply. 

Either F-5 or A-204 makes a satisfactory acid-hardening fixing 
.bath, 5 min immersion being sufficient. A silver recovery installa- 
tion is not practical for the scale of most phototemplate operations. 
It is the practice in this region to sell the exhausted fixing bath to 
local laboratories which recover the silver. These laboratories will 
make a periodic check to determine the silver content and the degree 
of exhaustion of the bath. Recovery is practical for them when the 
bath contains as low as 3 grams per liter, and as much as 25 grams 
per liter are recovered from some baths. 

All metal sheets must be backed with clear lacquer to keep them 
from reacting in the acid stop and fixing baths. 

Change of water in the wash tank should be rapid enough that 
washing is satisfactorily completed in 5 min. The processed ma- 
terials are squeegeed free of surface water and left in the room at- 
mosphere to dry. Drying time, 5 to 10 min for metal; 15 to 30 
min for reflex negatives. 

Metal sheets and other materials are moved and handled by hand 
in all operations except processing, where it is desirable to have a 
hoist on a monorail. The hoist assembly should be able to support 



142 F. DAVIS Vol 46, No. 2 

500 Ib and the monorail should run the short way of the tank group 
and extend well over either end of the tanks. 

There are various methods of fastening materials to the crane. 
Some workers punch holes in the tops of the metal sheets and insert 
hooks attached to a wooden bar, which carries a loop in the center into 
which the hook of the crane is caught. Others use small friction 
clamps at the tops of the sheets. One of the neatest devices is a hard- 
wood rod which has 2 vacuum cups mounted on it, and has a sturdy 
stainless steel foot attached at the bottom to take the weight of the 
material. At the top there is a handle for hand use and a slot for use 
with the crane. Two of these bars can be hung from a T-beam for 
large sheets. Glass reflex negatives up to 4 X 6 ft are placed in a 
hardwood frame which is suspended from a crane for processing. 

Trays. Smaller pieces of material are processed in trays of wood 
or stainless steel. In general the same formulas are used as for tank 
processing. 

In older installations employing the reflex method, A-79 in trays 
large enough to take 4- X 6-ft sheets has been used to great extent 
for developing reflex negatives to high contrast, but this developer in 
open tray has a life of only 4 to 6 hr. The recent introduction of re- 
flex paper of exceptional^ high inherent contrast made especially for 
line work has obviated the necessity for such contrasty and expensive 
developers. Consequently the longer lived and entirely satisfactory 
formulas A-103 and AP-80 are beginning to be generally used. The 
inefficiency and waste of large open tray installations for this method 
have long been recognized, and vertical tank installations are being 

made as rapidly as heavy production conditions permit. 



EXCITATION METHOD 

The X-ray method takes advantage of the fact that certain ma- 
terials absorb radiation at one level of energy and reradiate it at an- 
other. Aluminum sheets up to 5 X 12 ft are degreased, cleaned 
with wash thinner, primed with a clear lacquer, coated with 2 coats 
of white lacquer, and further coated with 10 coats of a cream-colored 
fluorescent lacquer, which on X-ray activation emits radiation in the 
short wave end of the visible spectrum to which silver halides are 
highly sensitive. This sheet is a drawing board on which the original 
engineering drawing is made. 

The fluorescent substance will not remain dispersed in the lacquer 
vehicle unless a pressure pot with agitator is used in spraying. Care 






Feb.. 1946 



SURVEY OF PHOTOTEMPLATE METHODS 



143 



must be taken that the highly abrasive fluorescent substance does 
not scour off too much of the pressure pot and agitator into a given 
batch of lacquer, otherwise the lacquer may become contaminated 
and deactivation result. These sheets should be left to dry by air 
overnight, after which they are sanded with fine sandpaper. 

All drawings were at one time made with 6H graphite pencil. 
Unavoidable variations in line widths in the originals made X-ray 
exposure quite critical so critical that there is now a strong tend- 
ency to abandon pencil in favor of a hypodermic needle pen and lac- 




FIG. 4. Consolidated Vultee's Downey plant designed and con- 
structed this X-ray printer for both primary and after-glow exposure 
procedures. It will accommodate 4 X 12-ft sheets. 



quer ink technique developed by Consolidated Vultee at Downey, 
California, which gives a line uniform in width and density. 

Original drawings in the aircraft industry are traditionally left- 
handed. A symmetrical shell on either side of an airframe axis is pre- 
supposed and only the left-hand half is drawn. Fluorescent master 
drawings can be reproduced either by a primary or an afterglow- 
primary method, depending on whether a right-hand or left-hand 
image is wanted. In general about 40 per cent of reproduction orders 
are for right-hand images so that primary exposure is all that is re- 
quired. In Consolidated Vultee's phototemplate laboratory at 
Downey, California, an original is placed face up in an X-ray printer 
containing 2 fruit inspection X-ray units operating at 100 kv peak 



144 



F. DAVIS 



Vol 46, No. 2 



and 7 ma. These are set at the bottom of a 6-ft concrete pit with 
6-in. side walls to absorb X-rays emitted at this potential. (See Figs. 
4 and 5.) One-inch plywood covers the pit and forms a table top, to 
which is hinged a lid carrying a vacuum blanket to insure good con- 
tact. A piece of nonfluorescent sensitized metal is laid emulsion 
down onto the original drawing, the lid lowered, and a 17-in. vacuum 
pulled in about 5 sec. After the lid, which is covered and skirted 
with lead to protect operators, is down a primary exposure of one 
^^^^^^^^^^^^^^^^^^^_ minute to X-rays is given. 

This is sufficient to produce 

.JJ^Jj a satisfactory latent image. 

All operations are controlled 
5^ by limit switches and inter- 

locking mechanisms to mini- 
mize hazard to operators. 
Right-hand images thus pro- 
duced have a black back- 
ground with white lines. All 
lettering is of course a mirror 
'image. 

The remaining 60 per cent 
of the orders are for left-hand 
images, which require metal 
intermediates corresponding 
to ordinary photographic 
negatives. Metal for inter- 
mediates is prepared in the 
same manner as the original 
drawing material. After 
the lacquer has dried and 
has been sanded it is sensitized with a silver halide emulsion. 
The intermediates are exposed by an afterglow technique in which 
the original engineering drawing alone is first placed face toward 
the activating units for an exposure of 5 to 10 min. During this 
time 'the electrons in the atoms of the fluorescent substance are 
rapidly pushed to outer orbits, but when the external energy source 
is cut off these electrons slowly drop back to their normal orbits, 
giving off energy and producing a phenomenon known as phosphores- 
cence, which frequently requires 6 to 8 hr to subside. 

After this activation, the original is placed face to face with the 




. The interior of Vultee's X- 
ray printer contains two 100-kv peak 
fruit inspection X-ray units. 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 145 

sensitized intermediate in a vacuum frame and held for 5 to 10 min, 
during which time the phosphorescence builds up a negative mirror 
latent image. The processed intermediate, or metal negative, is 
then printed onto sensitized metal stock by the primary exposure 
method to give a left-hand positive image. Both primary and after- 
glow images are developed in ordinary photographic solutions, fixed, 
washed, and dried. 

Flop-Matches. Since it is the practice in the engineering loft to 
develop only the left-hand half of a layout, when the shop requires a 
completed symmetrical layout about a centerline the phototemplate 
department must furnish it. This involves registering a right-hand 
image with a left-hand on the same sheet within the narrow toler- 
ances set by the department. To accomplish this a fluorescent 
metal negative is first made by primary exposure and from this a 
fluorescent metal positive is printed by the afterglow method. 
Small holes are drilled along the centerline of both of these at pre- 
cisely the same points. The metal negative is clamped to a large 
sheet of sensitized metal, into which another set of holes is drilled 
using those of the metal negative as a guide. The clamps are re- 
moved, the 2 sheets bolted together, and an exposure given by the 
afterglow method for the left-hand half. 

The metal negative is then removed, the metal positive bolted on, 
and an afterglow exposure given for the right-hand hah . The final 
image is a positive for the left- and a negative for the right-hand 
halves, but the lines are legible and hence acceptable to the shop. 

The excitation methods are size-for-size methods claiming "zero- 
zero size-change" since all printing is done by contact in vacuum 
frames (except for special jobs like "flop-matching") onto dimension- 
ally stable materials. 

Reference Reproductions. Negatives may be made 1:1 from 
fluorescent masters onto translucent vellum stock, and these printed 
onto sensitized translucent materials to make positives for the in- 
expensive production of reference Ozalid white prints requiring only 
2 steps. exposure and dry development. However, this scale limi- 
tation is a handicap in the case of original drawings 4 X 12 ft or 
larger, since sensitized translucent stock is at most 42 in. wide. 
Furthermore, reference drawings are unwieldly if too large. Conse- 
quently the use of a camera is indicated as an adjunct to the X-ray 
method for the reproduction of reduced scale reference drawings. 

Costs, A complete X-ray phototemplate laboratory installation 



146 F. DAVIS Vol 46, No. 2 

can cost anywhere between $15,000 and $100,000 depending on the 
method of working, the nature of the equipment, and the volume of 
work. 

For a time a phosphorescent lacquer method showed promise in 
this group since it required activation by only ordinary illumination, 
fluorescent tube, photoflood, or daylight, but in 2 years it was found 
that the phosphors completely refused to be reactivated. Some 
original drawings are used again and again for 10 years or more; 
fortunately the hundreds of boards that could not be reactivated 
were able to be reproduced by the reflex or camera methods. 

THE CAMERA METHOD 

Original engineering drawings are made on sheet' metal which has 
been coated with 2 to 4 coats of a matte lacquer with sufficient tooth 
to take 6H pencil or silver solder, and of sufficient plasticity that 
it will not chip. Silver solder is sharpened to a chisel point and 
because it is somewhat harder than graphite requires more pressure 
to make a line. This results in a slightly incised line which is not so 
easily smudged during the long-period development of a drawing on 
the loft floor. However, during such development lines are con- 
stantly changed, and because of this incision ordinary abrasive 
erasers soon dig through the lacquer coat to bare metal. It has been 
found that certain solvents such as trichloroethylene have a solvate 
action on some lacquer coatings, and if a cotton swab carrying a 
small amount of solvent is lightly touched to a silver solder line, it 
will disappear without leaving a trace. This results in a better 
quality drawing. 

It was at one time thought that drawing surface lacquer should 
be of a greenish or bluish tint so as not to cause eyes train among 
the draftsmen, but now nearly all engineering lofts are flooded with 
fluorescent illumination which causes no more eyestrain on a white 
board than on a tinted one; consequently, white loft boards are 
beginning to be used extensively because they give much more satis- 
factory photographic reproductions. 

Original drawings are placed on the vacuum copyboard of large 
cameras and copied onto negative materials following conventional 
photographic copying techniques. (See Fig. 6.) However, the need 
for accuracy is great, and no camera should be considered for the 
job which does not permit a controlled adjustment of 2 working 
planes to 0.0005 in. There are as yet no cameras on the market 



Feb., 1946 



SURVEY OF PHOTOTEMPLATE METHODS 



147 



especially designed for phototemplate work and such as are now in 
use have been adapted from the photomechanical field. Some of 
these have been modified to take metal sheets up to 5 X 12 ft. They 
are quite rugged in appearance and some of them weigh up to 5 tons, 
but in spite of this they are delicate precision instruments, and a 
repeated load of over 100 Ib on the -copyboard will throw them 
seriously out of line. It is often desired to project images onto 




FIG. 6. The Lanston monotype camera, which was designed for 
photomechanical requirements, has been adapted to the phototem- 
plate camera method. The vacuum copyboard is shown above. 



sensitized materials weighing considerably more than this, such as 
drill jigs and formblocks. Thus the cameras currently available 
have this shortcoming for the phototemplate method; however, if 
only lightweight metal sheet is used they are quite adequate. 

Glass plates are used for negative materials in the camera method 
and are of the high-contrast lithographic type. Hammer Ortho 
Offset and Eastman Kodaline have been found satisfactory for the 
purpose. Development is in a high-energy paraformaldehyde 
developer such as A-79 or Paralith, with stop bath and fixing bath 



148 F. DAVIS Vol 46, No. 2 

as usual. These plates admirably satisfy the requirement for high 
accuracy. Careful measurement of a 16-in. photographic line grid on 
both brands over a protracted period did not reveal any line dis- 
placement; accuracy of the 18-in. cathetometer used was 0.5 ju. 

The only size-change effects that can be expected from glass plates 
are those caused by temperature changes, the coefficient of expansion 
being about the same for the commercial plate that is used as that 
for mild steel. However, temperature changes in and about the 
camera are often extreme, particularly if a cool light is not used as a 
projection source, or if copy is heated excessively with arc lamps. 
Experience indicates that it would be economical in the long run to 
have the entire camera installation temperature controlled, thus 
eliminating the continued necessity for shooting test strips, or other- 
wise checking the accuracy of the image. The cost of temperature 
control ranges from $15,000 to $25,000. 

Scale of reduction onto glass negatives depends largely on job 
requirements. In the case of reproductions for ordinary airscrew 
models, photo department tolerances are 0.001 in. per ft, when it 
is practical to reduce the copy 4 to 5 times. At these ratios a 30-in. 
process lens will cover 5 X 10 ft satisfactorily at //45 and //32, re- 
spectively. There has always been a line-spread problem in the use 
of litho-type plates on pencil copy which results in large part from 
the fact that the pencil lines are not uniform in width nor density. 
The nature of the negative materials is such that if an exposure is 
given for the finer lines, there will be an excessive widening of the 
heavier lines. Cameramen have found that it is best to expose 
strongly in order to hold the heavier lines to a satisfactory width, 
and use Farmer's Reducer to clean out the finer lines during proc- 
essing. 

However, now that jet propulsion models are in production, toler- 
ances are only 0.003 in. in 10 ft, less than a third as much as 
formerly. In addition to the product demand for greater over-all 
accuracy, the insistence on better line quality is strong. As a result, 
reductions for these models are only 1:2. 

Templates are used in making tridimensional plaster patterns, 
from which molds are made for casting zinc or other alloy drop- 
hammer dies. These alloys shrink in various amounts depending on 
the alloy usually 0.10 in. to 0.125 in. per ft and allowances for these 
shrinkages can be easily made optically to produce templates oversize 
in the proper amount. The camera turns out such special jobs easily 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 149 

and quickly. However the alloys do not shrink uniformly, so that 
in spite of the optical corrections there is usually a lot of handwork 
required on the dies. This has led to the present tendency to break 
away from so-called "shrink" templates and get back to actual size 
templates. Hand grinding is still required, and check templates are 
used to check contour accuracy. 

The camera is used to a limited extent for the making of reduced 
scale templates for the construction of wind-tunnel models. 

Reference Reproductions. A most important use for the camera 
is in making reduced scale translucent vellum negatives from the 
originals. From these, translucent positives are made for the print- 
ing of the many thousands of Ozalid white prints required for refer- 
ence purposes in aircraft manufacture. No phototemplate de- 
partment can be complete without some kind of copying camera 
for this purpose. 

Flop-Matches. It is not satisfactory to turn a glass negative 
over image side away from the lens in order to project the right- 
hand half of a symmetrical contour even though customary com- 
pensations are made for refraction resulting from the thickness of 
the glass. Camera crews find that they spend hours trying to register 
the complex of fine lines at the centerline. This is because the amount 
of distortion, although zero on the optical axis, increases progressively 
as the extremities ot the field are approached. They have found it 
much more satisfactory to make a positive print on glass irom the 
negative. The glass negative with left-hand image including center- 
line is exposed and register points on the centerline developed up 
locally. The glass positive is then placed in the holder, registered on 
the developed centerline points, and exposed. The resulting match 
is half positive and half negative, but the image is accurate and the 
lines are legible. 

Costs. A complete phototemplate camera installation costs any- 
where between $40,000 and $150,000 depending on the nature of the 
equipment, the volume of work, and the job requirements. 

REFLEX METHOD 

Original engineering drawings are made with 6H pencil on sheet 
metal coated with white matte lacquer. There is no line-spread 
problem in this method and fine lines are reproduced faithfully. 
Therefore variations in line width and density do not cause trouble. 
The method employs common photographic contact printing pro- 



150 



F. DAVIS 



Vol 46, No. 2 



cedure. Even in the making of the negative, which is done by reflex 
printing, the materials are in contact. 

The negative material is prepared by applying a varnish-type 
adhesive to V^in. plate glass with a silk screen. The glass thus 
coated is placed horizontally in a rack for a sufficient time to allow 
the volatile solvents to escape about 45 min. It is then placed 
on a work table and a piece of reflex paper rolled out on the adhesive 
emulsion up. In one plant this negative, 3 X 5 ft to 4 X 6 ft, is 
immediately placed in a vacuum frame for a few seconds to push the 




FIG. 7. Making a reflex negative at Douglas Aircraft Company. 



paper tightly against the glass and force the adhesive into the paper 
fiber. In a long-period production test it has been found that by 
this method the negative can be exposed and processed within 30 
min without detectable size variation. However, it is good practice 
to allow sensitized glass negative stock to cure overnight at room 
temperature. The reflex paper is factory sensitized over high-grade 
paper stock with silver halide emulsion. Ordinarily paper for this 
purpose does not have a baryta coating, since maximum translucence 
is desired. It has been observed that for optimum reflex results the 
transmit ted-to-incident light ratio of the paper base is 1 :6; materials 
deviating appreciably from this ratio do not give good reflex images. 



Feb., 1946 



SURVEY OF PHOTOTEMPLATE METHODS 



151 



The printing box for this method is basically quite simple. A 
light-tight box is constructed large and sturdy enough to take a piece 
of heavy plate glass somewhat larger than the maximum size of draw- 
ing to be reproduced. One side of the glass is sandblasted to diffuse 
the light source; this side is placed toward the source, which consists 
of a bank of incandescent lamps so spaced as to give even illumina- 
tion. The intensity of the light is controlled by a variable resistance 
in the line, and is such that exposure times run 10 to 20 sec. (See 
Fig. 7.) 




Consolidated Vultee uses pin-located reflex negatives for the 
reproduction of large master drawings. 



Consolidated Vultee at San Diego has a printing table topped by a 
piece of VVin. plate glass 5 1 /* X 12V 2 ft to take 5 X 12-ft metal draw- 
ings. For making reflex negatives of a full 5 X 12-ft drawing, four 
3 X 5-ft glass negatives with a machined metal bushing set in each 
end are butted on long sides and firmly held in place on the printing 
table by tapered pins which pass into machined bushings set in regis- 
ter in the Y2-in. plate glass. (See Fig. 8.) 

Douglas Aircraft at Long Beach, California, has set up a printing 
table at an angle of 10 deg to the vertical with a l /2-in. plate glass 
surface 5Y 2 X 13 ft. For reproducing a 5 X 12-ft drawing, four 
3 X 5-ft sensitized glass plates are set up on end on a rigid base 
support and butted at the longer sides. No registering pins are used. 



152 F. DAVIS Vol 46, No. 2 

In either case the drawing is placed to face the sensitized glass 
negative surface, a neoprene vacuum blanket rolled over the sand- 
wich, a single exposure made, and the negatives developed. When 
these are dry they are placed in identically the same positions as 
when exposed, Y 4 -in. Scotch masking tape run down the butt joints, 
a piece of 5 X 12-ft sensitized metal placed to face the negatives, 
and a print made. Since such large prints are wanted in the shop 
for long contour developments, the three V4-in. gaps left by the 
masking tape are not objectionable. 

Orders for such large reproductions represent less than 5 per cent 
of total orders, most of them being for prints 4 X 6 ft or smaller. 
The method for these is the same as for the larger prints, without 
the necessity for registering or butting. 

The procedures just described produce only left-hand images. 
For a right-hand image a positive transparency is made onto glass 
negative stock from a negative. From this positive a right-hand 
image is made on sensitized metal or other material. 

The rationale of the reflex method is as follows. During negative 
exposure light is transmitted through both paper base and emulsion. 
When it strikes the white surface of the original drawing it is reflected 
back through the emulsion to the paper base. It continues to travel 
between these 2 reflectance layers until it builds tip a' strong latent 
image. Wherever the light strikes a graphite pencil line it is ab- 
sorbed, so that the final result on the negative is a clear line image 
against a black background. A good reflex paper should produce 
negatives on which the line image can be clearly seen at a distance 
of a few feet by reflected light, it should have a sufficiently high 
threshold that it can be handled for a short time in subdued daylight 
or room light, and should have high inherent contrast to produce 
clean sharp lines without fogging. 

Occasionally a red sensitivity in addition to yellow, green, and blue, 
is required in order to reproduce colors in proper value. This added 
sensitivity to red is desirable in those operations adopting the new 
hypodermic needle pen and lacquer inks, which are now being used 
in 6 colors including red. Colors are used for coding mold lines, 
pattern lines, centerlines, and the like, for immediate recognition 
on the loft floor; such information as is not required in the shop 
can be dropped in the reproduction process by proper filtering. 

Ordinarily where the reflex method is used reproduction planning 
is so efficient that all reproductions required from a given negative 






Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 153 

are made within 24 hr of receipt of the order, so that the negative is 
no longer needed. (See Fig. 9.) The plates are placed in large trays 
containing 10 per cent caustip solution and left to soak for several 
hours, when the adhesive is thoroughly softened and the paper with 
adhesive can be wiped off. The plates are washed and stood up to 
dry, when they are ready to be resensitized. 




FIG. 9. Where photographic images can be placed directly onto 
tool material, intermediate templates are unnecessary. Materials 
shown carrying engineering information are: 1-in. Richlite for form- 
block, V*-in. plowsteel for punch die, 0.036-in. galvanized steel check 
templates, positive and negative. 



Flop-Matches. A negative is made as usual from the left-hand 
information with centerline as furnished by engineering. Meantime 
a piece of 0.005-in. calendered unplasticized translucent Series V 
Vinylite large enough for the completed image is coated on both 
sides with a clear nitrocellulose lacquer. This sheet is then sensitized 
by brush on one side with emulsion containing butylcellosolve to 
make the emulsion adhere smoothly to the lacquer. An exposure is 
made on the sensitized Vinylite including centerline, a left-hand image 
developed, and the image dried. 



154 JF. DAVIS Vol 46, No. 2 

The other side of the Vinylite is then sensitized by brush, the 
developed center line, face up, is registered over the centerline on the 
negative, and another left-hand image exposed, developed, and dried. 
The result is a positive transparency with the entire symmetrical 
layout on it. This when printed produces white lines on a black 
background, the 0.005-in. thickness of the Vinylite not causing 
enough line fuzzing to give trouble. 

Reference Reproductions. Positive transparencies on sensitized 
materials can be made at 1:1 from the glass negative for the printing 
of Ozalid white prints. As with excitation methods this scale limita- 
tion is a handicap in the case of large original drawings. Here again 
some kind of copying camera is required for the reproduction of 
reduced scale reference drawings. 

Costs. A complete working reflex laboratory can be installed for 
from $3500 to $15,000, depending on the volume of work, the number 
and nature of the printing tables, and the job requirements. The 
$3500 unit is simple but efficient and can handle up to 3 X 5-ft 
negatives from any portion of a 5 X 12-ft original drawing. One 
large airframe plant has an installation costing only $6700 which for 
several years has consistently turned out an average of 26,000 sq ft 
of phototemplates per month of high quality and with great efficiency. 

OTHER METHODS 

Lithographic. The tooling department of one of Northrop Air- 
craft Company's plants at Hawthorne, California, has worked out a 
lithographic method for reproducing line images. A piece of trans- 
parent 0.010-in. plastic is laid over the portion of an original engineer- 
ing drawing that is to be reproduced, and fixed securely at the edges. 
Lines are scribed into the plastic with a sharp pointed tool. Scribing 
should not be too deep or the plastic will break. The burr is knocked 
off these lines, which are filled with black grease pencil. 

A piece of metal, lacquered black on both sides, is coated in a 
whirler with bichromated gum arabic. The whirler is a rotating 
table, usually horizontal, upon which the metal to be sensitized is 
centered. The sensitizing solution is poured onto the center of the 
metal and the rotation rate is such that the plate is coated uniformly 
to the edges. Only sufficient sensitizer to coat the area oi the plate 
is used. A circular band of metal is placed around the moving parts 
to prevent drops of sensitizer from flying into the room. Drying 
is hastened by electrical heating units placed in the hinged lid ot the 



Feb., 1946 SURVEY OF PHOTOTEMPLATE METHODS 155 

unit. Heat should not be excessive, since the sensitizer is hardened 
by heat as well as light. An exposure of 7 min is given to a 35-amp 
twin-arc bank through the transparent plastic sheet carrying the 
positive line image and other information, with the image toward 
the light. 

The plate is developed with a calcium chloride and lactic acid solu- 
tion and the lacquer removed from the unexposed lines with furfural. 
After washing, the plate is placed as anode in a tank of saturated 
sodium chloride solution and electrolytically etched for 1.5 min at 
800 amp, 7 v, d-c. The plate is washed again, dried, and coated 
with clear lacquer to prevent oxidation of the bright metal line image. 

No flexible transparent plastic sheet has yet been found that will 
hold size for long periods. Consequently great care must be taken 
in choosing material for the positive line image. Calendered un- 
plasticized transparent Vinylite Series V sheet 0.010 in. is satis- 
factory if printed within a few days of scribing, otherwise it must be 
trammeled against the original to check size. 

Electrolytic. One of Lockheed Aircraft Corporation's tooling 
departments uses an electrolytic method for reproducing line images. 
The method requires scribing through a nonfusible lacquer to the base 
of a sheet of body steel or galvanized steel. The layout is either 
made by hand from original engineering information, or a photo- 
graphic replica is furnished. The scribed image is moistened with 
transfer solution and placed face up in a press having thick pads of 
sponge rubber attached to the platens. A clean piece of terneplate, 
aluminum, or galvanized steel is placed on the scribed sheet. Copper 
bands across the rubber pads make contact with the 2 metal sheets. 
The press is closed and current applied for a few seconds. Pressure 
is released and the mirror copy, black line image on silvery back- 
ground metal, is washed and dried. To prevent oxidation on the 
clean metal surface, a coat of clear lacquer is usually applied. The 
method is fast, economical, and can turn out 100 to 200 copies before 
the lacquer of the original breaks down. 

CONCLUSION 

Table 1 gives a rough evaluation of the 3 phototemplate methods 
used most commonly in the West Coast region. 

It should be pointed out that no phototemplate department in the 
airframe industry necessarily uses one method to the exclusion of 
all others. In general, 5 years of heavy production experience have 



156 



F. DAVIS 



TABLE 1 

Comparison of the 3 Major Phototemplate Methods 

Excitation Camera 



Production, sq ft per month 
Photographic cost per sq ft 
Total personnel 

Personnel required with 5-10 years' photo- 
graphic experience 
Production, sq ft per man month 
Total Capital Investment 



29,000 

$0.74 

36 



None 

806 



$0.71 
62 

30 

516 



Reflex 



32,000 93,000 



$0.37 
56 

None 
1,660 



$80,000 $300,000 $50,700 



shown that where both a contact method for the accurate size-for- 
size reproduction of original engineering drawings and a copy camera 
for the production of intermediates for reduced scale reference 
drawings are used together, a complete and efficiently functioning 
phototemplate laboratory results. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 



American Cinematographer 

26, 10 (Oct., 1945) 

Telefilming Horse Races (p. 334) N. HARTFORD 

The "Guzap" That Went to War (p. 335) 
The Production of Scientific Films for Biological and 

Medical Purposes (p. 342) J. Y. BOGUE 

26, 11 (Nov., 1945) 
The History and Origin of 16 Millimeter (p. 376) A. F. VICTOR 

26, 12 (Dec., 1945) 
U. S. Government Report on Agfa Color Process (p. 

416) 
G. E. Develops Miniature Flash Tube (p. 435) 

British Kinematograph Society, Journal , 

8, 3 (July-Sept., 1945) 
Cutting Room Practice (p. 56) C. FREND 

British Kinematograph Society, Proc. Theatre Division 

(1944-45) 

The Post-War Kinema and Its Equipment Practi- 
cal Projection Problems (p. 2) R. H. CRICKS 
Double Reels and Film Mutilation (p. 11) S. WILLIAMS 
Kinema Screens and Their Maintenance (p. 14) H. C. STRINGER 
Over-all Projection Efficiency (p. 16) R. PULMAN 
Picture Presentation (p. 18) S. T. PERRY 
Kinema Engineering Efficiency (p. 19) H. E. WHITNEY 

Electronic Engineering 

17 (Dec., 1945) 
The Pye "Videosonic" Television System (p. 814) D. I. LAWSON 

Electronic Industries 

5, 1 (Jan., 1946) 

Lens Aberrations in Picture Projection (p. 86) A. MONTANI 

RCA Demonstrates Color Television (p. 178) 



157 



158 



CURRENT LITERATURE 



Vol 46, No. 2 



Institution of Electrical Engineers, Journal 
92 (Ft. 3), 19 (Sept., 1945) 
Studio Technique in Television (p. 165) 



International Photographer 

17, 9 (Oct., 1945) 

Production Control of Monopack (p. 7) 
The Dome (p. 18) 

17, 11 (Dec., 1945) 
8000 Pictures Per Second (p. 10) 
Glossary 3-Color Process Terms (p. 18) 
The Destycrane (p. 21) 
Television Topics (p. 22) 

International Projectionist 

20, 10 (Oct., 1945) 
Westrex Sound Systems (p. 12) 
Projectionists' Course on Basic Radio and Televi- 
sion Pt. 16, A.C. Circuits and Vectors (p. 18) 

20, 11 (Nov., 1945) 
The Projection Life of Film (p. 7) 
Step-by-Step Analysis of an RCA 16-Mm Amplifier 

(p. 12) 
Projectionists' Course on 'Basic Radio and Television 

Pt. 17, Resonance (p. 20) 
Color-Corrected Lenses Prevent Rainbow Effect (p. 

22) 

20, 12 (Dec., 1945) 
Introduction to Vacuum Tube Oscillator Circuits 

(P- 7) 

The Projection Life of Film (p. 12) 
The Stratovision System for Television, FM (p. 16) 

Motion Picture Herald 

161, 3 (Oct. 20, 1945) 
$5,000,000 Urged for Film Library (p. 44) 
Color Television Arrives; World Agreement Urged 
(p. 48) 

161, 7 (Nov.17, 1945) 

British Producers Demand Color Film Facilities (p. 
25) 

Photographic Journal (Royal Photographic Society) 

85B, 5 (Sept.-Oct., 1945) 
Television and the Kinema (p. 94) 
The Deposition of Metal Films : Their Application to 
Colour Photography (p. 97) 



D. C. BlRKINSHAW AND 

D. R. CAMPBELL 



W. J. KENNEY 
J. ALTON 

H. J. SMITH 
W. J. KENNEY 
J. ALTON 
W. S. STEWART 



G.S. APPELGATE 
M. BERINSKY 
R. H. TALBOT 
A. NADELL 
M. BERINSKY 



L. CHADBOURNE 
R. H. TALBOT 
C. E. NOBLES 



F. L. BURT 



P. BURNUP 



G. PARR 



J. YARWOOD 



Feb., 1946 



CURRENT LITERATURE 



159 



Photographic Society of America, Journal 

11, 10 (Dec., 1945) 

Edward J. Steichen (p. 527) G. E. MATTHEWS 

Cows, Movies and Color Photograph (p. 528) C. E. K. MEES 

Radio News 

34, 5 (Nov., 1945) 
Movies for Television (p. 10) S. PATREMIO 

34, 6 (Dec., 1945) 

Studio Acoustics Pt. 1, Basic Principles and Recent 
Developments Involved in Designing Broadcast 
Studios to Give the Most Desirable Acoustic Prop- 
erties (p. 5) R. H. BOLT 

35, 1 (Jan., 1946) 

Studio Acoustics (p. 8) R. H. BOLT 

Technique Cinematographique, La 
IS, 8 (Sept., 1945) 

An Automatic System of Phasing the Shutters of 
Camera and Projector in Background Work (p. 

135) L. MOHIER 
Color Motion Pictures by the Chimicolor Process (p. 

136) R. VALETTE 
Scanning Losses Caused by the Slit hi Motion Picture 

Sound Reproducers (p. 139) H. E. ROUSSIN 

15, 9 (Opt., 1945) 
New Negative Materials for Variable-Width Sound 

Recording (p. 151) D. O'DEA 

High Efficiency for Black-and-White and Colored 

Background Projection (p. 153) J. VERGENNES 

Conservation of Films (p. 154) 



1 



59th SEMI-ANNUAL TECHNICAL CONFERENCE 

Hotel Pennsylvania, New York 
May 6-10, 1946 

Directory of Committee Chairmen 

Atlantic Coast Section and Local Ar- 
rangements FRANK E. CAHILL, JR., Chairman 

Registration and Information W. C. KUNZMANN 

Luncheon and Dinner-Dance Commit- 
tee E. I. SPONABLE, Chairman 

Hotel and Transportation O. F. NEU 

Membership and Subscription Commit- 
tee JAMES FRANK, JR., Chairman 

Ladies Reception Committee Hostess MRS. O. F. NEU 

Papers Committee BARTON KREUZER, Chairman 

C. R. DAILY, Vice-Chairman 

Publicity Committee HAROLD DESFOR, Chairman, assisted by 

LEONARD BIDWELL 

Projection Programs 35-mm H. F. HEIDEGGER, Chairman, assisted 

by Members New York Projection- 
ists Local 306 
16-mm J. E. STEOGER 

HOTEL RESERVATIONS AND RATES 

No hotel room reservation cards will be mailed to the membership for this Con- 
ference. Therefore, members and others must book and reserve desired room ac- 
commodations early and direct with Joseph Troise, Front Office Manager, Hotel 
Pennsylvania, New York 1, N. Y., prior to April 20. Mention that the reserva- 
tion is in connection with the SMPE Technical Conference. No rooms will be 
assured or available unless confirmed by the hotel management. 

160 



FIFTY-NINTH TECHNICAL CONFERENCE 161 

Note: Out-of-town members who can schedule their New York arrival for 
Sunday, May 5, are more apt to get immediate room assignment on this date 
than if arrival is on Monday, May 6. 

The following per diem room rates, European plan, are extended to SMPE 
members and guests when booking accommodations direct with the Hotel Penn- 
sylvania : 

Room with bath, one person $3 . 85, $4 . 40, %4 . 95, $5 . 50, $6.05, $6.60 

Room with bath, 2 persons, double bed $5. 50, $6.05, $6.60, $7. 15, $7.70 

Room with bath, 2 persons, twin beds $6.60, $7. 15, $7.70, $8.25, $8.80 

Parlor suits for one or 2 persons $10.00, $11.00, $13.00, and $18.00 

REGISTRATION 

The Conference registration headquarters will be located on the 18th floor of 
the hotel adjacent to the Salle Moderne, where all business and technical sessions 
will be held during the 5-day Conference. Members and guests are expected to 
register. The fee is used to defray Conference expenses. 

TECHNICAL SESSIONS 

SMPE members and others contemplating presentation of papers at this 
Technical Conference can greatly assist the Papers Committee in the early as- 
sembly of the program by mailing in title and author of papers together with an 
abstract by April 1, 1946. Complete manuscripts should be sent to the Chair- 
man or Vice-Chairman of the Papers Committee not later than April 15. 

Only through your earnest cooperation will it be possible to draft and announce 
the papers program prior to the opening of the Conference. 

SMPE GET-TOGETHER LUNCHEON 

The Society will again hold its regular pre-war social functions, and accordingly 
a Get-Together Luncheon is scheduled in the Penn Top (formerly the Roof Gar- 
den) on the 18th floor of the hotel, on Monday, May 6, at 12 : 30 P.M. Ladies are 
invited to attend this luncheon. Tickets must be procured at the registration 
desk prior to noon on May 6, so that adequate hotel accommodations may be pro- 
vided accordingly. 

The Board of Governors cordially invites the holders of Dinner-Dance tickets 
to spend a social hour with the Board in the hotel Georgian Room between 7 : 15 
P.M. and 8:15 P.M., on May 8, preceding the Conference dinner. (Refreshments.) 

The informal Dinner-Dance (dress optional) will be held in the Georgian Room 
promptly at 8: 30 P.M., on May 8. Dancing until 1 : 30 A.M. 

Note: It is imperative that Dinner-Dance tickets be procured and table reser- 
vations made at the registration headquarters prior to noon on May 8. Your 
earnest co-operation with the Arrangements Committee is requested. 

LADIES' PROGRAM 

A reception parlor will be provided in the hotel for the ladies' daily get-together 
and open house. The ladies' entertainment program will be announced later. 



162 



FIFTY-NINTH TECHNICAL CONFERENCE Vol 46, No. 2 

MOTION PICTURES AND RECREATION 



Conference identification cards issued to registered members and guests will 
be honored at New York deluxe motion picture theaters which will be listed in later 
issues of the JOURNAL. 

Those interested in other entertainment while in New York should consult the 
hotel information bureau, or the SMPE registration headquarters. 



Scheduled, 



9:30 a.m. 



12 :30 p.m. 



2: 00 p.m. 
8: 00 p.m. 



9:00 a.m. 

9: 30 a.m. 
2:00 p.m. 



Monday, May 6, 1946 

Open Morning. 

Hotel, 18lh Floor: Registration. Advance sale of Luncheon and 

Dinner-Dance tickets. 
Hotel Penn Top: (formerly Roof Garden) , 18th Floor: Get-Together 

Luncheon. (Eminent Speakers.) Note: Luncheon tickets must 

be procured before noon on May 6, at the registration desk. 
Salle Moderne: Opening session of the Conference. Business and 

Technical Session. 
Salle Moderne: Evening Session. 



Hotel, 18th 

tickets. 

Salle Moderne: 
Salle Moderne: 
Open Evening. 



Tuesday, May 7, 1946 

Floor: Registration. Advance sale of Dinner-Dance 



Morning Session. 
Afternoon Session. 



Wednesday, May 8, 1946 

Open Morning. 
10:00 a.m. Hotel, 18th Floor: Registration. Advance sale of Dinner-Dance 

tickets. 

2:00 p.m. Salle Moderne: Afternoon Session. 
7:15 p.m. Georgian Room (Reception Foyer): A social hour with your Board 

of Governors preceding the Dinner-Dance. (Refreshments.) 
8:30 p.m. Georgian Room: Fifty-Ninth Semi- Annual Technical Conference 
Dinner-Dance. Social get-together, entertainment, and dancing 
until 1:30 A.M. 

Note: Tickets must be procured and tables reserved prior to noon on 
May 8, for this function. 



Thursday, May 9, 1946 

Open Morning. 



2:00 p.m. 
8: 00p.m. 



Salle Moderne: 
Salle Moderne: 



Afternoon Session. 
Evening Session. 



Feb., 1946 FlFTY-NlNTH TECHNICAL CONFERENCE 

Friday, May 10, 1946 



163 



9:30 a.m. Salle Moderne: Morning Session. 

2: 00 p.m. Salle Moderne: Afternoon Session. Adjournment of the Fifty- 
Ninth Semi-Annual Technical Conference. 

Note: All sessions during the 5-day Conference will opeh with an interesting 
35-mm motion picture short. 

IMPORTANT 

Those desiring hotel rooms must book their accommodations direct with the 
Hotel Pennsylvania management prior to April 20, which are subject to cancella- 
tion prior to May 1. 

Owing to the acute travel conditions, it is imperative that out-of-town members 
and guests who contemplate attending the May Technical Conference consult 
their local railroad passenger agent regarding rail and Pullman accommodations, 
within the existing Pullman reservation period. 

W. C. KUNZMANN 
Convention Vice-P resident 



SOCIETY ANNOUNCEMENTS 
ATLANTIC COAST SECTION MEETING 



The first meeting in 1946 of the Atlantic Coast Section of the Society was held 
on January 16 when Rollin W. King and Emmanuel Berlant discussed the Robo- 
tron camera. Mr. King dealt with the scope of the equipment as it relates to medi- 
cal, dental, and scientific photographic records. Mr. Berlant described the engi- 
neering aspects of the camera. 

"Specialty photographic equipment usually found on the market is either too 
cumbersome or difficult to operate," Mr. King said, "and generally requires an 
operator with photographic background and considerable experience. Therefore, 
many photographic records are not made that would be of great value in the vari- 
ous professions. Most equipment used hi the professional fields medical, dental, 
scientific, etc. has been adapted from commercial or amateur models, which 
often prove inflexible. The Robotron camera was designed to fill the require- 
ments of these specific fields." 

Mr. Berlant described the engineering development of the camera, designed for 
use by the government during the war. "It combines in one lightweight unit a 
self-contained high-speed vapor discharge flash tube, with complete automatic 
operation. It is necessary only to focus and frame the picture, all settings for 
shutter speed, diaphragm stop, etc., being eliminated," Mr. Berlant said. 

Swinging Into Step, a 16-mm motion picture furnished through the courtesy of 
the Army Pictorial Service, opened the meeting held at the Hotel Pennsylvania, 
New York. 

PACIFIC COAST SECTION 

The Pacific Coast Section of the Society opened its 1946 series of programs with 
a meeting on January 22 at which W. Bradford Shank, of the Federation of 
Atomic Scientists, spoke on "Atomic Energy and Civilization." The speaker's 
highly interesting presentation was followed by an open forum in which questions 
from the audience on all phases of the subject were discussed. 

The meeting, held in the ERPI Review Room, opened with a showing of the 
Army Air Forces motion picture, The Last Bomb. 



We are grieved to announce the death of William B. Bamford, Asso- 
ciate member of the Society, on April 10, 1945, in Belmar, New Jersey. 
164 



SOCIETY ANNOUNCEMENTS 165 

MEMBERS LOST SERVING THEIR COUNTRY 

The Society desires to compile a list of members who gave their lives while 
serving with the Armed Forces of their country. Such a list will include members 
abroad who served with Allied military forces as well as those in the services of 
the United States. 

The general office of the Society is not always advised of deceased members, 
and it will be appreciated if readers of the JOURNAL will forward the name of any 
member known to them to have been a war casualty. Please include with names 
submitted the approximate date, place, and any other information available. 

Your co-operation will assist the general office in obtaining a complete and 
accurate list for the records of the Society. 



EMPLOYMENT SERVICE 

POSITIONS OPEN 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 



Position available for Optical Designer, capable of handling the calcula- 
tion and correction of aberrations in photographic and projection lens 
systems. Junior designers or engineers will be considered. Write 
fully giving education, experience, and other qualifications to Director 
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi- 
cago 45, m. 

POSITIONS WANTED 

Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. J. K., 
354 Ninth Ave., New York 1, N.Y. 



Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 



Newsreel cameraman, overseas U. S. Army veteran with honorable 
discharge, desires position with educational or commercial organization 
with work-training arrangement. Will supply prints on Army work in 
ETO. For full references, experience, and record, write Tom J. Ma- 
loney, 406 Oak St., Ishpeming, Mich. 



SOCIETY of MOTION PICTURE ENGINEERS 

MOTCL PENNSYLVANIA NCW YOftK I, N-Y- Ttt. PCNN. 6 O62O 

APPLICATION FOR MEMBERSHIP 

(This page should be completely filled out by applicant in conformity with Qualifications and Re- 
quirements given on the opposite page for grade desired. References given should be members or 
nonmembers who will supply information on applicant's experience and serve as sponsors.) 

Name__ Age 

Address : 

City , 

Employer_ 



Occupation. 



Grade Desired: Associate Q ; Active 
Education * 



Record of Employment* (list companies, years, and positions held). 



Other Activities*. 



REFERENCES 

(Name) (Address) (City) 



The undersigned certifies that the statements contained in this application 
are correct, and agrees, if elected to membership, that he will be governed by the 
Society's Constitution and By-Laws so long as his connection with the Society 
continues. 

Date 19 (Sgd) . 

* If necessary, use additional sheet to give complete record. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 MARCH, 1946 No. 3 

CONTENTS 

PAGE 

An Application of Direct-Positive Sound Track in 16- 
Mm Release Processing by Duplication Method 

G. C. MlSENER AND G. LEWIN 167 

The Wartime Record and Post- War Future of Projec- 
tion and Sound Equipment A. G. SMITH 178 
The Measurement and Control of Dirt in Motion Pic- 
ture Processing Laboratories 

N. L. SIMMONS AND A. C. ROBERTSON 185 
A New Recorder for 16-Mm Buzz Track 

M. G. TOWNSLEY 206 
New Permanent Magnet Public Address Loudspeaker 

J. B. LANSING 212 
Specialized Photography Applied to Engineering in the 

Army Air Forces P. M. THOMAS AND C. H. COLES 220 
Colored Trace Oscillograms 

L. S. TRIMBLE AND F. W. BOWDEN 231 
Current Literature 237 

Fifty-Ninth Semi-Annual Technical Conference 239 



Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCIETY of MOTION PICTURE ENGINEERS 

MOTCl PENNSYLVANIA NCW YORK I. N'Y- TfL. PCNN. 6 O62O 

HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 

* President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 

* Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
^Executive V ice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
^^Engineering Vice-President: JOHN A. MAURER, 
37-01 31st St., Long Island City 1, N. Y. 

* Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
** Financial Vice-President: M. R. BOYER, 
350 Fifth Ave., New York 1. 

* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 

* Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
^Treasurer: EARL I. SPONABLE, 
460 West 54th St., New York 19. 

Governors 

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
**WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
"JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairman, Atlantic Coast Section. 
**Term expires December 31, 1947. JChairman, Pacific Coast Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 March, 1946 No. 3 

AN APPLICATION OF DIRECT-POSITIVE SOUND TRACK 

IN 16-MM RELEASE PROCESSING BY DUPLICATION 

METHOD* 

G. C. MISENER AND G. LEWIN** 



Summary: When dupe negatives are employed to make large releases on 16- 
mm positive, 3 successive printing operations are normally involved in arriving at 
the release print. That is, the master positive, dupe negative, and release print are 
made in order. Since each printing operation is attended by some degree of quality 
degeneration, it is desirable to reduce the operations to a minimum. 

In the application described, the first printing is eliminated by rerecording to direct 
positive on 16-mm film. The amplifier channel feeding the direct-positive recorder is 
suitably equalized and compressed for 16-mm projection conditions, and bridges the 
channel used to make the normal rerecorded negative for 35-mm release. A 32-mm 
release negative is then made on a nonslip printer from the 16-mm direct positive. 

Introduction. The 16-mm release prints of most Signal Corps 
films have been made by a laboratory which employs the dual 32-mm 
dupe negative process in making volume release. The usual pro- 
cedure, when working from a 35-mm rerecorded negative, is to make a 
35-mm fine-grain master positive, reduce the fine-grain master to 
2 tracks on the 32-mm dupe negative, and release print on 32-mm 
positive stock which is subsequently slit. Each of the 3 printing 
and developing operations thus involved contributes some degree of 
image deterioration. Close control of each operation is necessary in 
order to avoid cumulative errors in sensitometric conditions or 
printing quality, which might lead to serious distortion and poor in- 
telligibility in the final print. Moreover, a troublesome amount of 
noise may result from the handling involved in each stage of the 
process, particularly that caused by foreign material and abrasion. 

It might reasonably be suggested that the most satisfactory method 

* Presented Oct. 16, 1946, at the Technical Conference in New York. 

* Signal Corps Photographic Center, Long Island City, N. Y. 

167 



168 G. C. MlSENER AND G. LEWIN Vol 46, No. 3 

of eliminating certain of these printing steps would be to extend the 
application of rerecording. That is, it would be possible to rerecord 
directly to the 32-mm release negative simultaneously with the 
making of the 35-mm rerecorded negative, through a parallel channel 
arrangement. Another possibility would be to rerecord "electric 
dupes" from a nonslip print of the 35-mm rerecorded negative, as is 
common practice in making 35-mm release. 

These solutions to the problem were not feasible, however, since a 
special recorder for the 32-mm stock was unavailable. This being 
the case, use of a direct-positive 1 recorder, to permit rerecording 
directly to the fine-grain master positive, suggested itself. This 
would not only eliminate one stage of printing and developing, but 
would also afford the opportunity to introduce equalization and com- 
pression in the 16-mm rerecording channel suitable for 16-mm pro- 
jection conditions. The latter was a most important consideration 
as has been established by experience in recent years. 2 - 3 

In the cases of films which will have both 35-mm and 16-mm re- 
lease, the previous practice at the Signal Corps Photographic Center 
has been to make only one rerecorded negative, using frequency and 
compression characteristics, which represented compromises between 
the optima for 35-mm and 16-mm release. Lack of sufficient equip- 
ment, as well as time and film economy considerations, precluded 
making two 35-mm rerecorded negatives, with special characteristics 
for the 2 classes of release. 

Equipment. A 16-mm recorder was selected for conversion to 
the direct-positive arrangement, a unit capable of high-quality per- 
formance being on hand. Use of 16-mm stock would also affect an 
economy in raw stock. The recorder, a Maurer Model D, was con- 
verted by the manufacturer to make direct-positive variable-area 
track of the bilateral type. An unmodulated bias line of approxi- 
mately 5 mils is normally used. The conversion included, of course, 
modification of the galvanometer to expose track of standard print 
width and placement. 

The recording and noise-reduction amplifiers used with the re- 
corder are standard Maurer units, Models 120-B and 130- A, respec- 
tively. Along with the amplifier and recording lamp supply, they 
are mounted semipermanently on a rack near the recorder. A Maurer 
16-mm film phonograph is located on the same bench with the re- 
corder, as shown in Fig. 1. Since the recorder and film phonograph 
are equipped with 1800 rpm synchronous motors permanently at- 



March, 1946 APPLICATION OF DIRECT-POSITIVE SOUND TRACK 169 

tached, interlock operation was provided for by coupling interlock 
motors to the shafts of the synchronous motors with bonded rubber 
couplings. The interlock motors are camera type with the gear 
reduction modified for 2400 to 1800 rpm operation. 

The electrical arrangement of the complete 16-mm channel and its 
relation to the 35-mm channel are indicated in the block schematic 
of Fig. 2. Mounted on the racks of one of the 35-mm rerecording 
channels are the components which complete the 16-mm channel. 
This section of the 16-mm channel consists of a bridging coil, a vari- 




FIG. 1. Layout of 16-mm equipment. 

able T-pad, an RCA MI-10206-C electric mixer or compressor, an 
MI -10108 variable high-pass filter and a variable compression ceiling 
control pad. The setting of the latter affects the amount of com- 
pression realized, once the recording amplifier gain has been set at a 
suitable fixed value. The channel components appear on patch bay 
jacks and are normalled together. Thus, to set up for parallel re- 
cordings on 35-mm negative and 16-mm direct positive, it is merely 
necessary to patch the bridging coil across the bridging buss of a 
35-mm channel. The MI-10206-C compressor has an attack time of 
approximately 0.001 sec, and is adjusted for a compression rate of 
approximately 2:1. 



170 



G. C. MlSENER AND G. LEWIN 



Vol 46, No. 3 



Experience has shown that it is generally good practice to compress 
the over-all signal, with the volume range reduced to as little as 15 db 
for 16-mm projection conditions. 2 ' 3 The intelligibility of speech, 
especially the low level passages, is enhanced, and low level music and 
effects are made audible above ambient projection room noise. Per- 
haps the one undesirable result is that high level music and high level 
effects, such as gunfire and bomb explosions, are held down, whereas 
the effects, at least, might normally be permitted to overshoot the 
modulator considerably. The MI-10108 high-pass filter is adjustable 
to sharp cutoff positions at 80, 100, 120, 135, and 150 cycles, and has 
an off or flat position. 




FIG. 2. Block schematic of parallel recording arrangement. 

Recording Characteristics. Comparison listening tests, made 
with standard issue 16-mm projection equipment set up in a fairly 
large room, have indicated that the low-frequency cutoff may well be 
in the range of 100 to 120 cycles. This cutoff reduces masking of 
intelligibility by the reverberation of reproduced low frequencies, 
without seriously affecting the content of low frequencies in music 
and effects. It was found that the high-frequency cutoff could be as 
low as 5000 cycles without appreciable detriment to the quality of 
reproduction. A low cutoff frequency was favored, since it would 
exclude the higher frequencies which would require more critical 
control of sensitometric conditions and printing quality to minimize 
cross-modulation or envelope distortion. 



March, 1946 APPLICATION OF DIRECT- POSITIVE SOUND TRACK 171 

The listening tests also indicated a satisfactory amount of com- 
pression to be 14 db into 7 db. That is, the ceiling control is so ad- 
justed that the 14-db range of input immediately below the 100 per 
cent modulation level is compressed into 7 db. The compressor 
input pad is adjusted so that average peaks modulate between 80 
per cent and 100 per cent on the 16-mm track. Hence, the top 25 db 
of volume range on the 35-mm dubbing prints are compressed into 
approximately 18 db on the 16-mm direct positive. This means that 
even the lower levels of signal will produce at least 10 per cent 
modulation on the 16-mm track, insuring a satisfactory signal-to- 
noise ratio. 




5 100 3 5 1000 

FREQUENCY 
FIG. 3. Frequency characteristic of 16-mm channel. 



10000 



The voice modulation will be maintained at a particularly good 
level since, as we noted in the schematic, a compressor is also em- 
ployed in the speech position of the rerecording console, ahead of the 
point bridged by the 16-mm channel. The mixer bases his deter- 
mination of the compression used in the speech position on his judg- 
ment of the 35-mm channel monitor quality. In the instances of 
original narration tracks recorded with 20 into 10 compression, little, 
if any, compression is applied in recording. As much as 20 into 10 
is used if the original track has no compression. 

The frequency characteristic of the 16-mm recording channel, from 
bridging coil input to galvanometer input, for 3 positions of the high- 
pass filter, is shown in Fig. 3. The response of the galvanometer is 



172 



G. C. MlSENER AND G. LEWIN 



Vol 46, No. 3 



essentially flat over the recording range covered. The rise shown in 
the high-frequency region results from the film loss equalization in the 
recording amplifier. The combination of this equalization with the 
low-pass filter leaves a broad peak in the range which contributes most 
to intelligibility, namely, 2500 to 5000 cycles. However, the rise 
in this range is sufficient to provide the desired film loss and intelli- 
gibility equalization only when combined with the film loss equaliza- 
tion in the photocell amplifiers of the 35-mm rerecording channel. 
The over-all characteristic from 35-mm photocell amplifier input to 




DB 



100 



10000 



5 1000 2 

FREQUENCY 

FIG. 4. Over-all 35-mm and 16-mm recording characteristics; photocell 
amplifier input to modulators. 



both 35-mm and H)-mm rerecording busses is shown in Fig. 4. As 
indicated, a relatively extended frequency range is used for the 
35-mm release, with the low-frequency cutoff at 45 cycles for music 
and effects, and approximately 100 cycles for speech. The low-pass 
filter is usually set at 8000 cycles. 

Although provision has not yet been made to allow the mixer to 
switch his monitor to the 16-mm channel for quality checks, the 
generally fixed relationship between the 35-mm and 16-mm recording 
characteristics rather obviates the necessity for such an arrangement. 
Moreover, the contrast in monitor quality between the 2 channels 
might tend to affect the mixer's critical evaluation of the 35-mm 
characteristic. A good quality monitor speaker is provided in the 



March, 1946 APPLICATION OF DIRECT-POSITIVE SOUND TRACK 173 

16-mm machine room, and since this room is acoustically treated and 
the recorder is quiet in operation, the recordist is able to detect ab- 
normal quality in the 16-mm channel. 

Film Characteristics. In order to compare the film losses in- 
volved in the 2 methods of duplication, a frequency run was recorded 
on both the 35-mm and 16-mm machines. The 35-mm negative and 
16-mm direct positive thus obtained were used as the originals in the 
processes of duplicating and printing. The resulting equivalents of 
release prints, as well as the originals, were measured on soundheads 
which are capable of giving reliable modulation level indications at 




5 7 100 23 57 1000 23 57 IOOOO 

FREQUENCY 

FIG. 5. Over-all system and film characteristic as measured on 16-mm re- 
lease print. 

the higher frequencies. In each case, calibrated frequency test films 
were also measured in order to determine and deduct the losses of the 
soundheads and their associated systems, and thereby obtain the 
actual characteristics of the films in question. 

It was found that the printing losses sustained in the direct positive 
to release print process were somewhat less than those incurred in the 
old procedure involving 3 printing operations. For example, approxi- 
mately 2-db improvement at 6000 cps was noted for the direct- 
positive method, while at 4000 cps the difference was one db. 

The image losses in the 2 originals were also determined. This 
was done by comparing the effective modulation values obtained on 
the soundheads with the impressed amplitudes on the tracks as meas- 



174 



G. C. MlSENER AND G. LEWIN 



Vol 46, No. 3 



ured with a microscope. The 16-mm direct positive has somewhat 
greater image loss, as might be expected, since, for example, 6000 
cycles on 16-mm track is the equivalent in image width of 15,000 
cycles on 35-mm film. The 6000-cycle loss of the 35-mm negative was 
found to be less by approximately 2.5 db. Thus, it is seen that the 
total film losses for the 2 methods, from modulator to effective release 
print modulation, are about equal, the lower image loss in the 35-mm 
negative approximating the lower printing losses in the direct-positive 
process. 




1.0 1.2 1.4 1.6 1.8 2.0 

RELEASE PRINT DENSITY 

FIG. 6. Cross-modulation of release prints. 



2.2 



It is emphasized at this point that consideration of the average 
relative film losses, particularly since they are essentially equal, is of 
secondary importance. Elimination of the possible variations, image 



16-Mm Direct Positive 
32-Mm Dupe Negative 
16-Mm Release Print 



TABLE 1 

Sensitometric Conditions 



Stock 

5372 
5302 
5302 



Total 
Density 

1.40 
2.22 
1.48 



Base 
Plus Fog 

0.28 

0.40* 

0.04* 



Density 
Latitude 



1.35-1.60 



Gamma 

2.30 
2.60 

2.60 



Includes print-through. 



March, 1946 APPLICATION OF DIRECT- POSITIVE SOUND TRACK 



175 



losses, and handling effects in a stage of printing and processing, is 
the more important consideration. 

An actual over-all frequency characteristic realized under normal 
operating conditions, i. e., the photocell amplifier input to effective 
16-mm release print modulation characteristic, is shown in Fig. 5. 

Sensitometric Conditions. The stock used for the 16-mm direct- 
positive recording is Eastman Type 5372, a high contrast fine-grain 
emulsion on blue base. The characteristics of this film have been de- 
scribed in the JOURNAL previously. 4 A blue filter, Jena No. BG12, is 




1.2 1.4 1.6 1.8 2.0 

RELEASE PRINT DENSITY 

FIG. 7 Cross-modulation of release prints. 



employed in the recorder to enhance image resolution. The labora- 
tory makes the 32-mm dupe sound negatives on fine-grain positive 
stock with a nonslip printer and UV exposure. The release prints 
are made on sprocket- type printers, using fine-grain positive stock 
and white-light exposure. A set of sensitometric conditions which 
gives satisfactory results is tabulated in Table 1. Other com- 
binations of dupe negative and release print densities will also give 
good results, as will be seen in Fig. 6. The optimum sensitometric 
conditions were determined by making cross-modulation tests and 
plotting the results against various density parameters. A 4000- 



176 



G. C. MlSENER AND G. LEWIN 



Vol 46, No. 3 



cycle carrier, amplitude modulated approximately 75 per cent at a 
400-cycle rate, was used as the test signal in these investigations. 

Although there would ordinarily be no occasion to reproduce the 
direct-positive itself on a soundhead, its cross-modulation cancella- 
tion was found to be approximately 20 db at the total density value 
of 1.40, or a silver density of 1.12. Still greater cancellation is at- 
tained at lower but less practical density values. While the density 
contrast of the dupe negative, which is printed from the low-density 




1.2 



1.6 



1.7 



1.3 1.4 1.5 

RELEASE PRINT DENSITY 

FIG. 8. Density tolerance for 30 db cross-modulation cancellation. 



direct positive with its blue base, is not as high as might seem de- 
sirable, the density contrast and effective modulation of the release 
print are good. In fact, higher values of release print density are 
indicated for this method than is the case when duping from the 
regular 35-mm rerecorded negative. At the same time the clear 
area density is only about 0.04. The data of Table 2 demonstrate 
that the effective release print modulation level is good over a range 
of workable sensitometric conditions. 

The family of curves in Fig. 7 represents the cross-modulation re- 
sults obtained when printing dupe negatives of D2.22 and D2.42 
to a range of release print densities. Essentially, the same data are 



March, 1946 APPLICATION OF DIRECT-POSITIVE SOUND TRACK 177 

plotted in another form in Fig. 8 to show the density tolerances in 
the direct-positive and release prints, for dupe density of 2.22. It 
will be noted that reasonable ranges of densities give satisfactorily 
low cross-modulation. 

TABLE 2 

Effective Modulation Level at 400 Cycles for Dupe Negative Density of 2.22 

Reproduced 
Density Level 

Direct Positive 1 . 40 2.0 

Direct Positive 1 . 56 -2.0 

Release Print 1 . 48 0.0 

Release Print 1 . 30 0.0 

Z52.8 Test Film ... 0.0 

All 3 sound tracks have same amplitude. 

REFERENCES 

1 DIMMICK, G. L., AND BLANEY, A. C.: "A Direct-Positive System of Sound 
Recording," /. Soc. Mot. Pict. Eng., XXXIII, 5 (Nov., 1939), p. 479. 

2 BRIGANDI, P. E.: "Rerecording 35-Mm Entertainment Films for 16-Mm 
Armed Forces Release," 7. Soc. Mot. Pict. Eng., 44, 1 (Jan., 1945), p. 18. (In- 
cluding appended discussion by E. M. Honan.) 

3 MILLER, W. C.: "Preliminary Report of Academy Research Council Com- 
mittee on Rerecording Methods for 16-Mm Release of 35-Mm Features," /. Soc. 
Mot. Pict. Eng., 45, 2 (Aug., 1945), p. 135. 

4 O'DEA, D.: "Comparison of Variable- Area Sound Recording Films," /. 
Soc. Mot. Pict. Eng., 45, 1 (July, 1945), p. 1. 



THE WARTIME RECORD AND POST-WAR FUTURE OF 
PROJECTION AND SOUND EQUIPMENT* 

ALLEN G. SMITH** 



Summary. This paper is a case history of the manufacture and distribution of 
35-mm motion picture projection and sound equipment under wartime restrictions as 
imposed by War Production Board regulations. It tells about the difficulties and 
problems which were encountered and overcome so that the Armed Forces might ob- 
tain all of the equipment needed, with a limited quantity for our civilian theaters. The 
paper also discusses the reasons why our manufacturers cannot now produce as much 
new equipment as will be needed for replacements, and makes recommendations for 
future planning. 

Every person in the motion picture industry will remember the 
uncertain status of all manufacturers and distributors of projection 
and sound equipment, repair parts, and accessories soon after the 
Pearl Harbor disaster. We had a war on our hands. 

The War Production Board in Washington, D. C., was the one war 
agency charged with the responsibility of conserving vital materials, 
labor, and plant facilities so that the Armed Forces could have full 
and uninterrupted access to the productive capacities of every 
industry. Motion picture projection and sound equipment, like 
many other commodities so necessary to our commercial and indus- 
trial development, quickly became a wartime casualty. Certain con- 
servation and limitation orders issued by the War Production Board 
prohibited the manufacture of projection and sound equipment ex- 
cept that which was needed to fill orders for the Armed Forces. 

At the beginning of World War II our military minds worked 
quickly and thoroughly to overcome the difficult problems of training 
the staggering number of men needed for combat duty. They did 
not have enough skilled instructors available for the training of such 
large forces of men who were inducted into the different armed serv- 
ices. Naturally, their attention was drawn to the use of motion 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
* * National Theater Supply, New York; formerly Chief, Theater Equipment 
Section, War Production Board, Washington, D. C. 
178 



FUTURE OF PROJECTION AND SOUND EQUIPMENT 179 

picture projection and sound equipment, from which was created 
many ingenious synthetic training devices. 

Some of our manufacturers were called upon to work with the 
Ordnance Department of the U. S. Army, and the Department of 
Special Devices, Bureau of Aeronautics of the U. S. Navy, in the de- 
velopment of many different types of synthetic training devices 
through the use of the standard types of equipment which each 
manufactured. These amazing devices, after having been put into use 
in the training programs, saved millions of dollars in ammunition 
and guns and countless lives in the vigorous training campaigns. 

The Armed Forces needed standard types of equipment for other 
training purposes. Army training centers and Navy shore stations 
were provided with equipment for specialized training courses in all 
phases of military activity. The equipment was also used for the 
showing of films produced especially for the Medical Corps illus- 
trating proper surgical, first aid, and other techniques. Aside from 
the technical uses of the equipment, motion pictures were the only 
form of entertainment available in many military installation and 
training centers. They had a morale job to take care of and ' 'movies' ' 
played a most important part. 

The reputation of our manufacturers for the production of precision 
parts spread to the Army and Navy material procurement staffs. 
The Army and Navy could not produce in their own plants the preci- 
sion equipment and instruments required. Consequently, they had 
to turn to plants with known reputations for the production of pre- 
cision equipment. The production of motion picture equipment for 
the most part requires precision manufacturing within tolerances 
of Vio,ooo of one inch and many manufacturers were given contracts 
for certain high precision combat instruments. 

This meant expansion of plant facilities, the installation of new 
and intricate machine tools, and the training of operators to handle 
the tools. And yet, we could not stop production of our standard 
types of sound and projection equipment because the Armed Forces 
also needed that class of equipment. 

Industry manufacturers had a difficult problem to crack in getting 
as much equipment as required because of the high priority war work 
they were doing on direct war material. It then became the job of 
the War Production Board to correlate the activities of all the manu- 
facturers of projection and sound equipment into one co-ordinated 
function. If the Armed Forces needed projection and sound equip- 



180 A. G. SMITH Vol 46, No. 3 

ment, they needed it complete in every detail, even though many of 
the different units were being manufactured in different plants in 
many different cities. Their program could not be carried out if 
everything was ready for installation except the lamp houses, or the 
sound system, or even a pair of lenses. Everything had to be de- 
livered on time. The War Production Board scheduled the production 
of every unit of equipment needed by the Armed Forces in such a 
manner that all units arrived at their destination on time, notwith- 
standing that the manufacturing cycle extended from 3 months on 
lamp houses to 12 months for sound equipment. 

Because of the .demands of the military on the productive capacities 
of the equipment manufacturers for their normal product and mili- 
tary items, the motion picture industry faced a tremendous task in 
keeping its theaters in operation. The production of almost every 
item of equipment needed had been stopped, including projection 
and sound equipment, etc. Repair parts for sound and projection 
equipment could not be purchased because production of them had 
likewise been stopped. 

What could be done to relieve the distress of the motion picture 
industry? The answer could not be readily reached because all of 
our industrial resources were geared to war production. We hardly 
dared to suggest a solution which, in any manner, might interfere with 
the delivery of military goods. Yet, here was an industry which by 
its limitless energy was selling more war bonds, collecting more 
money for the Red Cross, and doing a bigger job of prosecuting the 
war effort generally than any other single industry. Copper, rubber, 
paper, tin, and aluminum were collected. Millions of dollars were 
spent through the War Activities Committee to make each co-oper- 
ating venture a success from which the only return was the satisfac- 
tion that the industry was doing its level best to help win a war. 
These accomplishments could not be continued unless the theaters 
could continue to operate. They needed repair parts, accessories, 
and a limited amount of new equipment to make the more urgent 
replacements of equipment which could no longer be economically 
repaired or efficiently maintained. 

The industry tightened its belt and learned for the first time the 
true meaning of maintenance, improvising, and substituting. Theaters 
patched and reversed their carpets, spent countless hours welding and 
nursing their auditorium seats. Some of their customers steamed 
and sweated in the summer because air-conditioning plants could not 



March, 1946 FUTURE OF PROJECTION AND SOUND EQUIPMENT 181 

be repaired or Freqn was not available. Some patrons practically 
froze in their seats for two winters because fuel was not available, 
or the heating plant could not be repaired. But the industry operated 
just the same because it' still had pictures on the screen and they did 
talk! 

The supply dealer soon became the theater owners' best friend and 
confidant because he still had a few repair parts and a little new equip- 
ment on hand in the early months of the war. If he did not have the 
necessary parts, he would rush emergency equipment to the theater. 
The supply dealer was his own rationing board so far as his rapidly 
dwindling stockpile of new equipment was concerned, and he par- 
celed out that new equipment only when its installation became the 
factor between closing down or continuing to operate a theater. He 
took pride in his part of the job, in return for which he received the 
overwhelming thanks of the theater owner. 

By April, 1942, the importance of the motion picture was firmly 
established with the top officials in Washington. The War Activities 
Committee of the Motion Picture Industry had been organized, the 
leadership of which was vested in the biggest and most resourceful 
men in our business. They brought the efforts of the industry to the 
attention of men like WPB's Donald Nelson, Secretary of the Treas- 
ury, Henry Morgenthau, Jr., and Secretary of Commerce, Jesse 
Jones. President Roosevelt knew what was happening in our busi- 
ness. The industry received immediate recognition of its efforts and 
the Amusement Section of the War Production Board was subse- 
quently established, later changed to the Theater Equipment Section. 

In August, 1942, R. J. O'Donnell of the War Activities Committee 
asked me to accept an appointment to the War Production Board as 
a Consultant on problems relating to the production and distribution 
of motion picture equipment in general, which appointment was ac- 
cepted with humility and a deep sense of responsibility. 

Our immediate problem was the production of repair parts for 
projection and sound equipment. In October, 1942, the Iron and 
Steel Order and the Copper Order were each amended to provide for 
the production of repair parts. Priority regulations, however, made 
it necessary at that time for theater owners to furnish their supply 
dealer with a priority rating for delivery of a part. We needed ac- 
cessories, such as rewinders, reels, change-over devices, film cabinets, 
etc. We needed a limitation order which would permit the produc- 
tion of repair parts and accessories, and which could be carried in 



182 A. G. SMITH Vol 46, No. 3 

stock by supply dealers and sold without priority ratings. We needed 
a WPB instrument by which we could schedule production and regu- 
late distribution of a limited quantity of new equipment then urgently 
needed to replace worn-out or burned-out equipment. 

Because of these circumstances, the WPB deemed it desirable and 
necessary that jurisdiction over the sound and projection equipment 
industry should be placed in one Industry Division, and that control 
over the production and distribution of projection and sound equip- 
ment be co-ordinated into one limitation order. Therefore, Limita- 
tion Order L-325 was prepared and issued so that the necessary de- 
mands of the Armed Forces and of the civilian theaters might be met, 
and yet to provide that there shall be no interference in the produc- 
tion of actual war goods. 

Under the terms of that order enough new equipment was scheduled 
for manufacture to make urgent replacements in civilian theaters. 
Dealers' stocks of repair parts and accessories were gradually re- 
plenished and sold without restrictions. 

The credit for the orderly and continued operation of the motion 
picture theaters belongs to no one individual. The success of our 
business during the war years can be attributed to a combination of 
ideals and ideas contributed by the War Activities Committee, the 
manufacturers and distributors of projection and sound equipment, 
the Society of Motion Picture Engineers, the projectionists, the 
officials of the War Production Board, and countless others. 

Now that the war is over doubtlessly most of us face the immediate 
future production of sound and projection equipment with undue 
optimism. There are 17,000 theaters in the United States, which 
figure represents only a small percentage of the theaters in the world. 
Our manufacturers must not only now plan to produce enough new 
equipment to meet a pent-up demand in the United States of 4 years, 
but they will also be called upon to fill orders for equipment to re- 
habilitate theaters in bombed-out areas and for thousands of new and 
modern theaters which will be built in nearly every foreign country 
in the world. 

Naturally, our manufacturers will build equipment to fill domestic 
orders first, at least enough to make the immediate and urgently 
needed replacements, and for new installations. Some new equip- 
ment will be exported because it is good for our business to have 
American made projection and sound equipment in use in foreign 
theaters. Our manufacturers must protect their foreign markets. 



March, 1946 FUTURE OF PROJECTION AND SOUND EQTjrPMENT 183 

It may also be expected that some of the better used equipment which 
is replaced by new equipment in the United States will be rebuilt and 
restored for export, thereby making more new equipment available 
for domestic installations. 

If we look at cold and practical figures, we get a startling picture of 
the job our manufacturers face. We require in the United States 
annually approximately 4500 standard projectors, 2250 standard 
sound systems, 6200 arc lamps, 4500 rectifiers, and 1000 motor-gen- 
erator sets. We also need nearly 1500 of the portable types of 35-mm 
sound projectors. Multiply these figures by the pent-up demand of 
4 years, and we have a need in the United States for 18,000 projectors, 
9000 sound systems, 24,800 arc lamps, 18,000 rectifiers, and 4000 
motor-generator sets. We will need 6000 of the portable types of 
sound projectors. Only approximately 50 per cent of one year's pro- 
duction has been manufactured under wartime restrictions for non- 
military uses since 1942, and considerably fewer motor-generator sets 
were produced since 1942. 

How can our manufacturers meet this demand for new equipment 
within a reasonable length of time? It is not an easy job for several 
reasons. Before the war, when materials and components flowed 
freely, the production cycle ranged from 3 to 10 months, depending 
on the class of equipment produced. That can be cut in the near 
future because new and better machine tools will be available. Our 
manufacturers have the "know-how" of mass production of precision 
instruments, a technique acquired during the war years. 

The time when our manufacturers will be delivering equipment in 
large quantities will depend greatly upon the availability of parts and 
components which they must buy elsewhere. They now must wait 
for other manufacturers to convert their facilities to meet our own 
specifications for such necessary parts as ball bearings, fractional 
horsepower motors, indicating meters, a host of electronic parts, 
switches, etc. Foundries must be geared to make the precision cast- 
ings we now need. Lumber in large quantities will be needed for 
shipping boxes. It will take time to train the additional skilled labor 
needed for such expanded production. 

Theater owners should consult their supply dealers and evaluate 
their equipment requirements now. The dealer was most helpful 
during the war years, and you may need his friendly co-operation and 
expert advice now more than ever. He might suggest the installation 
of a third projection unit, an emergency rectifier, or motor-generator 



184 A. G. SMITH 

set, or dual amplification for your sound system. The demand for 
so much new sound and projection equipment now would not be so 
great if such installations had been made before the war. It is insur- 
ance well worth the additional costs, and theater owners should look 
forward to the future when new equipment will be available in greater 
quantities so that such emergency equipment can be purchased. 

Until new equipment is generally available to all who need it, we 
should not relax our efforts to keep in perfect repair the equipment 
we must continue to use for the time being. The theater patron de- 
mands perfect projection and sound, and by continuing to keep our 
present equipment properly maintained, we reduce the basic causes 
of dangerous and costly film fires, and thus keep faith with the public. 



THE MEASUREMENT AND CONTROL OF DIRT IN 
MOTION PICTURE PROCESSING LABORATORIES* 

N. L. SIMMONS** AND A. C. ROBERTSONf 



Summary. The commonest sources of contamination of motion picture film 
in processing laboratories are discussed, and suggestions are made for preventing 
foreign matter finding its way onto the surface of the film. The "tacky-dish" technique 
and other methods of detection of dirt sources are described, and examples are given of 
surveys made in actual Hollywood laboratories. 

The film handled in processing laboratories can be contaminated by dirt associated 
with the outside air used in the laboratory, by personnel, by mechanical equipment, or 
by processing solutions. It has been found that when there are complaints about dirt, 
etc., a systematic investigation usually reveals the source of the trouble quickly. 
This is particularly true if quantitative measurements ar'e taken. These data allow 
one to construct a flow sheet which indicates the place where dirt is entering the 
system. The new tools used for this purpose are the smoke recorder, the "tacky-dish, 1 ' 
and the method of "lifting" samples. The Greenough microscope is a most useful and 
almost indispensable adjunct. 

The choice of air filtration systems and the proper kinds of uniforms are also dis- 
cussed briefly. 

Introduction. The superior screen quality of present-day film 
is the result of much care in the processing laboratories. Dirt is an 
ever-present enemy of screen quality and is one of the main obstacles 
to further progress now that sensitometric quality has reached such a 
high level. 

The methods of locating and minimizing sources of dirt have been 
developed to a high degree by the film manufacturers. When one 
observes what harm can be wrought by "loose dirt" it can be under- 
stood that dirt, firmly anchored in wet emulsion, would present a 
hopeless problem. The dirt control techniques which have been de- 
veloped by the manufacturers to aid them in the handling of film 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
* * Eastman Kodak Company, Hollywood. 

t Department of Manufacturing Experiments, Eastman Kodak Company, 
Rochester, N. Y. 

Figs. 1-4 reprinted from Industrial and Engineering Chemistry, 13 (May 
15, 1941), by permission of the American Chemical Society, copyright owner. 

185 



186 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

have been used by some photographic processing laboratory managers 
to advantage, and more recently, by various war industries. The 
latter include manufacturers of electrical equipment, radio tubes, and 
optical devices. The sources of contamination can be: 

(1) Outside air which generally carries with it unwanted material in the form 
of cinders, dust, and smoke. 

(2) Personnel, who introduce dirt into the process by reason of their activity 
and clothing. 

(5) Equipment, which may have rust or oil on it, or metal grindings resulting 
from the wear of moving parts. 

(4) Processing solutions, because of sediment, sludge, and scums. 

The problem is to learn which sources are dominant. 

During the past 2 years, 3 Hollywood laboratory superintendents 
have requested that a thorough survey of their plants be made with 
the idea of locating and eliminating the outstanding sources of dirt. 
In each case, investigation disclosed the existence of unsuspected 
causes for the presence of dirt on the negative or positive film. Like- 
wise, it was found that the original ideas held by the laboratory super- 
intendents as to the likeliest sources of dirt in their respective plants 
were not always verified by the survey. Each laboratory took 
steps to eliminate the specific dirt sources and to improve general 
conditions. Retests made by the "tacky-dish" and other techniques 
(to be described later) indicated that the gain in cleanliness of the air 
was considerable and that the improvement in the product was well 
worth the effort expended. 

Air Filtration. The drying cabinet of a motion picture develop- 
ing machine is fundamentally a likely source of dirt on motion picture 
film, since at this location the emulsion is wet and tacky and is exposed 
to a large mass of air. For this reason, it is primarily essential that 
the air used for the drying operation be filtered. It is also of great 
benefit to cleanse by filtration all air entering the film processing 
laboratory. In order to facilitate the entrance of clean air only, a 
slight positive pressure should be maintained inside the laboratory so 
that open doors, etc., act as clean air exits rather than inlets for un- 
cleaned air. 

Air-borne dirt may be classified according to size. The coarsest 
class includes dust, cinders, and lint-like particles. Smoke is much 
finer and has many particles smaller than 0.5 mu in size. The coarser 
dirt particles can be removed from air easily by air washers using 
sprays of water, or by viscous filters. This latter type includes a host 






March, 1946 MEASUREMENT AND CONTROL OF DlRT 187 

of appliances which depend for their effect upon the adhesive action 
of liquids of high viscosity held upon supports of the most diverse 
sorts. The objective in designing an air filter of this kind is to pro- 
duce a support of relatively low air resistance and large area. The 
designers have produced many forms of filters in their attempt to 
meet these requirements. Among the viscous filter supports com- 
monly used are ones made of crimped wire, folded paper, wire mesh, 
and glass fibers. 

Filter media of the viscous type may be of the "throw-away" 
variety for small installations, or the permanent cleanable types 
where there are enough units to justify a cleaning operation and sub- 
sequent reoiling. Where air is used in large volumes for noncritical 
operations, the "moving-curtain" automatic self -cleaning filter is 
cheap and effective. This filter consists of a device which supports 
and drives an endless curtain which is slowly drawn through a bath 
of oil held in the base of the filter. This operation cleans the filter 
continually and minimizes the need for close attention. The manu- 
facturers' recommendations should be followed in the installation of 
viscous air filters since too low an air velocity gives poor efficiency. 
An air velocity which is too high leads to spraying of oil, and can be 
very serious. 

Filters using bags made of canton flannel have frequently been 
used in filtering air. They do a better job of air filtration than vis- 
cous filters but they have enough drawbacks so that their installation' 
cannot be generally recommended. The fabric bags can generate 
lint when they are improperly installed so that they rub together. 
In addition, they often dislodge showers of dirt after they have been 
shutdown temporarily. Collapsing the bags loosens the dirt, which is 
free to move when the fans are started. Filter bags also have the de- 
fect that their capacity is badly affected as they remove dirt from the 
air. Their resistance often increases very rapidly at the end of their 
useful life. 

Filters made up of bundles or mats of soft crepe paper have been 
used. In one form the mats of paper are held in book-like frames of 
iron mesh which are then assembled in sets of 3 or 5 pairs to form a 
pocket-type filter. The pocket units are then assembled into headers 
in the usual fashion. Under some circumstances, it seems best to use 
these filters at lower than recommended face velocity. 

Electrostatic filters are more efficient than the foregoing types. 
They remove most of the smoke in the air, commonly working at an 



188 



N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 



efficiency above 95 per cent. Smoke is removed only to the slightest 
extent by viscous filters, their efficiency in this respect generally 
being 5-10 per cent. However, it is not always necessary to remove 
most of the smoke from air unless one desires to lengthen intervals 
between duct cleanings. No air filter is completely efficient, there- 
fore, some smoke remains in the air which has passed through it. 




FIG. 1. Smoke recorder open for examination. 



While smoke is made up of particles so small that they settle very 
slowly indeed by the effect of gravity, nevertheless they do deposit 
from air. Such smoke particles, or dirt, are slowly deposited on the 
walls of the duct by impingement and other causes. Later, the dirt 
is dislodged by vibration, or similar disturbances and flies out of the 
duct in a shower. Accordingly, it is good practice to place a viscous 
filter at the end of a long run of pipe in order to stop such showers of 
dirt. Smoke caught in electrostatic filters can also be dislodged upon 
occasion, Electrostatic filters of the self -cleaning type have the 



March, 1946 



MEASUREMENT AND CONTROL OF DIRT 



189 



attractive feature that their efficiency is not badly affected by failure 
of the high-tension electrical system. 

The efficiency of air niters with respect to smoke is measured by a 
smoke recorder 1 (Fig. 1). The smoke recorder functions on this 
principle: Dirt is drawn into an evacuated container through a small 
slit (Fig. 2) and impinges upon a piece of paper moved past the jet 
by a 24-hr clock. The air stream, which is moving with the speed 
of sound, deposits the dirt it is carrying upon the paper, where it 
can later be measured (Figs. 3 and 4). The instrument is quanti- 
tative since the jet delivers a constant volume of air as long as the 



U I |5 INCH J IDENT. NO. 



KODAK SMOKE 
RECORDER JET 




TAPER ^ INCH PER FOOT 
FIG. 2. Jet for smoke recorder. 

pressure difference between the 2 sides is greater than 0.53 atmos- 
phere. The 2 dark lines constituting the record are measured. 
Since one line represents the raw air and the other line the cleaned air, 
the efficiency of the filter can be calculated from these data. 

The instrument is easily portable and makes continuous records. 
It is of great assistance in studying air filter installations to see if the 
performance is as good as the design intended. If a departure is ob- 
served from the values one would have predicted from laboratory 
studies, then one must look about. It is often found that there are 
leaks around the edges of the filter housings, or that the fan is in- 
stalled on the clean air side of the filter and that leaks in the ducts, 
etc., allow dirty air to be drawn into the system. 



190 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

The gravimetric measurement of coarser particles is done according 
to the practices of the American Society of Heating and Ventilating 
Engineers by drawing measured samples of air through an alundum 
crucible filter, which is later weighed in order to learn how much dirt 
has been collected. By measuring the dirt concentration in the 
filtered and unfiltered air, the efficiency of the filters can be com- 
puted. 

All air ducts should be inspected regularly even when there are 
filters in the system. Accumulation of very fine particles on the 
duct surfaces may give a constant supply of black agglomerates of 




FIG. 3. Charts on holder for visual examination of dirt traces. 

soot and other atmospheric dirt. This can be avoided by regular 
cleaning of all ductwork. 

The use of conditioned air of from 60 per cent to 70 per cent relative 
humidity will minimize the pickup of dirt by the film as the static 
tendency of the film is at a minimum at these conditions. 

Contamination from Personnel. While it is comparatively easy 
to produce clean, lint-free air, it is difficult in practice to avoid diffi- 
culties caused by lint since the clothing and movement of workers 
continuously produces a steady cloud of fine particles of various sorts 
unless appropriate steps are taken to prevent the occurrence. The 
nature of the process dictates the steps that must be taken. In film 
manufacture, the workers have special uniforms adapted to the proc- 
ess involved. Men working near freshly coated film wear new uni- 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 191 

forms which are changed daily. Dandruff is a serious source of con- 
tamination, so the uniforms include head coverings which confine the 
workers' hair. Loose particles are prevented thereby from falling on 
the work, and the workers are prevented from getting hair, oil, etc., 
on their hands and thence on the work. It has been found that raw 
edges or threadbare fabrics are the sources of much lint. After 
certain indications of wear, the uniforms are used in less critical 
areas. These uniforms finally are used by the departmental me- 
chanics, whose need for lint-free clothing is not great. 

Workers in other departments, such as those where film is proc- 




FIG. 4. Instrument for photoelectric examination of dirt traces. 

essed, are normally uniformed only to the extent of wearing smocks 
and caps. However, these workers are forbidden to wear woolly, 
fluffy sweaters. The latter are notorious for the way they spread 
fibers broadcast. 

Not all the lint comes directly from the clothing of workers; they 
may dislodge it from other locations by their movement. Accord- 
ingly, working surfaces should be wiped frequently with a wet cloth. 
Floors should be mopped, never cleaned by being swept with an 
ordinary broom. In order to prevent generation of dust from chipping 
or surface disintegration, floors should be waxed with a soft, dirt- 
retentive wax. Under some conditions, a soft varnish using much 
tung oil can be employed. Concrete needs frequent attention in this 
respect. 



192 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

If the best operating conditions are desired, one must use a "clean- 
cab," which resembles a chemical fume cabinet, except that the air 
runs backward, and is carefully filtered. These arrangements have 
been made familiar by the advertisements of penicillin plants and 
optical factories. The use of clean-cabs is a utilization of a general 
scheme of routing dirt as one would route product. If this concept is 
borne in mind, the organization of precautions becomes a more rea- 
sonable matter. 

Carefully filtered air is useless unless it is used while still clean; 
therefore, it is sent over the work, past the worker (who is an im- 
portant source of dirt), and then back to the filter. Since lint is the 
thing to be removed, the air need not be sent back to the central 
station, but can be filtered and recirculated by a local unit using a 
viscous filter. Personnel bring in dirt on their shoes. In many loca- 
tions the entrance is narrow, thus forcing the worker to walk over a 
number of cocoa mats, etc. The mats remove much of the street dirt, 
but if more care is needed, an automatic shoe cleaner can be em- 
ployed. 

Designation of clean areas in the laboratory should stop all unneces- 
sary traffic of personnel and hand trucks through rooms where film is 
exposed and handled. 

Dirt from Equipment. Equipment made of materials which will 
serve well enough for most manufacturing purposes generally proves 
disappointing in the photographic industries. Slight amounts of 
corrosion, for example, mean little in most devices. The usual cri- 
terion is "how long will it last?" However, in photography the 
question is, "how long will it stay clean?" For example, some hard- 
used equipment will look nice if painted white, yet after a while the 
wear and tear of daily use will chip off the paint. Trouble ensues 
either from flakes of paint, or rust from the unprotected metal sur- 
face thus exposed. Such machine surfaces can be advantageously 
protected by heavy nickel- or chrome-plating. Ordinary plating is 
not adequate. 

Surfaces which rub together generate dirt, and provision must be 
made to localize the contamination. This contamination may be 
caused by over-generous lubrication or, in the case of neglected bear- 
ings, metal powder. Difficulty from this source is more widespread 
than one might think since many surfaces rub together. It is usual 
to think of bearings only in terms of shafts alone. For example, 
door latches and hinges fall in the category of moving parts. 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 193 



i 



Other causes of dirt are to be found in the work rooms themselves. 
Concrete floors have been mentioned and comments made about 
methods for preventing dusting as a result of surface disintegration. 
Battleship linoleum is a good floor covering if there is no heavy 
trucking and if special care is taken to hold the edges down by 
cementing and the use of mop-boards, etc. Walls should be smooth 
so that dirt will not lodge freely, and solid so that dirt cannot be 
shaken off easily. In critical locations it sometimes pays to cover 
plaster or concrete walls with canvas to prevent dirt from sifting from 
cracks. 

Since many of the minus-density spots found on prints are located 
near splices, special attention should be paid to splicing benches. Re- 
winding operations can collect dirt because of static electrification. 
This effect is generally small and is made smaller by the use of rela- 
tive humidities above 50 per cent. Electrical conductivity of film 
increases rapidly with humidity, and allows static charges to leak 
off. However, splicing operations generally permit the film to touch 
the worktable. This contact should be kept to a minimum by taking 
care not to overrun, etc., in rewinding. It is good practice to work 
on tables of white glass, which are intrinsically clean and which, by 
reason of their color, cause dirt to be easily visible. 

Combating dirt from the actual splicing operation itself is a prob- 
lem. One can easily avoid dust and prevent clippings, dried cement, 
etc., from accumulating, but one cannot easily handle the scrapings 
which come from a properly sharpened scraper blade. Properly 
placed suction will do much to remove the film scrapings from the 
work area. This device is used successfully by a number of labora- 
tories. Dirt from splicing is primarily a matter of care on the part 
of the operator, and demands understanding and cooperation from 
every splicing machine operator for its prevention. 

Many processing laboratories use cloth-covered drums in the drying 
cabinets for buffing the base side of film. Often the cloth wrapping on 
these drums becomes a lint source because raw edges of cut cloth may 
be allowed to flap against the film as the drums revolve. A little care 
in proper sewing of the cloth sleeves for covering the drums will 
usually take care of this dirt source. Proper choice of a lintless fabric 
is helpful in this connection also. 

White gloves are used by splicing machine operators, cleaning room 
workers, and others in laboratories in order to prevent fingerprints. 
If these gloves are given a thorough laundering prior to use, much of 



194 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

the lint will be removed.* It is often observed that workers using 
such gloves allow them to become exceedingly dirty, to such a point, 
in fact, that they introduce dirt. It is a good plan to provide a closed 
receptacle for soiled gloves, and a container for a supply of clean 
gloves, in order to maintain orderliness and to make it easy for work- 
ers to discard gloves too dirty to use on film. 

One common method of cleaning film is to rewind it, usually in a 
room set aside for this purpose alone, through a soft cloth pad satur- 
ated with cleaning solvent. The solvent employed is either carbon 
tetrachloride or a petroleum fraction of high volatility. It has been 
found that oftentimes there is a large concentration of lint in such 
cleaning rooms. Plush pads are notoriously bad for their release of 
lint fibers unless they are used with care and skill. The accumulated 
load of dirt is generally dropped from the pad at each stop unless care 
is taken to prevent such action. Pads should be made by folding the 
plush and sewing it so that the cut edges are enclosed within the pad. 
Suction vents may be supplied adjacent to the sources of generation 
of lint. This is highly desirable also as a means of removing toxic or 
inflammable solvent vapors. 

Another source of loose dirt is that caused by film rubbing sidewise 
against guides during rewinding. 

Processing Solutions. Experience in the field of apparatus 
construction is great enough that extensive compilations of recom- 
mendations are available. 2 The properties of various materials are 
given, as well as methods for fabricating them. 

Formation of calcium sulfite in developing solutions can bring about 
many attendant difficulties, such as scale formation on rollers, scum 
on film, and sludge in the solutions. This can best be prevented by 
filtering the solutions. For this purpose various types of cloth 
filters are prepared. Diatomaceous earth as a filter aid is used in 
a number of laboratories for obtaining greatest clarity of developing 
solutions. 

Other means of preventing the formation of insoluble calcium salts 
in developers are (1) the use of softened water, and (2) the use of cal- 
cium-sequestering agents, such as sodium tetraphosphate 3 in the de- 
veloper formulas. 

Some motion picture laboratories mix developers and do not filter 



* Gloves treated with synthetic resin particles dispersed in water have shown 
themselves to be nearly lint-free. 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 195 

them, but allow a period of standing for the calcium sludge to floccu- 
late and settle out. Sedimentation continues for a long period of time 
owing to the slow formation and settling rate of very fine crystalline 
particles of calcium sulfite and other similar compounds. 

The need for cleaning and polishing motion picture negative follow- 
ing development would be much reduced if the water-insoluble sub- 
stances were removed from the processing solutions. Likewise, the 
maintenance cost of the developing machines would be much re- 
duced, because of avoidance of scale build-up on the rollers, shafts, 
and other parts. 

A microbiological slime often forms in wash- water tanks.' The 
growth of the slime can be prevented by the use of sodium penta- 
chlorphenate or similar compounds. If slime becomes established, 
thorough cleaning is necessary, which may include circulating a warm 
solution of trisodium phosphate through the parts of the circulating 
system which cannot be reached with brushes. The concentration 
of trisodium phosphate needed depends upon circumstances and may 
be from 1 /z per cent to 3 per cent. 

Measuring Dirt. In order to study and control the unwanted 
material we call dirt we must be able to describe its occurrence 
quantitatively. Measurements not only allow us to gauge improve- 
ments, but generally furnish the information that makes improve- 
ments possible. Knowing the direction of air travel and relative dirt 
concentrations, one can follow the trail to its beginning. Dirt in the 
air is measured by such methods as : 

(Z) Sedimentation, 

(2) Filtration, 

(5) Centrifugation, 

(4) Impingement, 

(5) Thermal precipitation, 

(6) Electric precipitation. 

Three simple methods of dirt evaluation have been used very suc- 
cessfully in locating the chief sources of dirt in motion picture labora- 
tories. These methods are, (a) the tacky-dish technique, (b) the 
examination of samples lifted by adhesive tape, and (c) the micro- 
scopic examination of the film itself. Various related techniques are 
discussed by Bloomfield and Dallavalle, 4 and by Drinker and Hatch 5 
in their publications, which also include extensive bibliographies. 

It is important, in undertaking a complete study of dirt sources in a 
motion picture laboratory, to lay out first in detail the handling pro- 



196 



N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 



cedure, step by step, for both the negative and print, so that the in- 
vestigational methods may be applied in an orderly fashion and at 
the most critical points in the laboratory. 

The particular techniques to be described are useful primarily in 
locating concentrated localized dirt sources, which may cause more 
trouble than general personnel and building conditions. However, no 
amount of detective work done in locating sources of dirt can replace 




FIG. 5. Tacky dish used in dirty location. 

a continued insistence on high standards of maintenance and care on 
the part of personnel. 

Tacky-Dish Technique. This method is useful in measuring 
contaminants settling from the air, as well as contamination from 
equipment and personnel. It is particularly adaptable to the meas- 
urement of dirt carried in the film drying air. 

The method consists of exposing a definite area of a tacky surface 
(Fig. 5) to the atmosphere that is being tested for a known length of 
time. A solution is prepared by soaking one part of gelatin in 7 parts 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 197 

of cold water for 30 min, followed by warming to approximately 
135 F until all the gelatin is dissolved. Glycerin, 2 parts, is added 
and a trace of thymol for preservative action. This solution is poured 
into the bottom of a glass Petri dish. When the entire surface has 
been covered by tipping, the dish the excess is allowed to drain off 
and the top of the dish is immediately put into place. Both top and 
bottom of the dish must be thoroughly clean and free from lint. 
This surface will remain tacky for a period of several days, which is 
ample for most tests. 

The covered dishes are uncovered at the point where the sample is 
to be taken. For most determinations the dishes are placed in a 
horizontal position so that the particles in the dish represent particles 
falling directly into the dish or settling out of the air. When placed 
directly in the air stream in developing machine drying cabinets, 
the test dish may be placed upside down, if necessary, on the wire 
grating which often covers the bottom of the drying cabinets, so that 
the air impinges directly on the tacky surface. It is possible to use 
. the dishes in a vertical position; this may be necessary when there is 
a right angle bend in an air duct and the cleanliness of the air in the 
duct is wanted. Dishes placed in front of and behind an air filter will 
give an indication of the filter efficiency for large particles by this 
method. 

When the sampling has been completed the dishes are covered care- 
fully and their contents examined microscopically. The Greenough 
type of binocular microscope is especially recommended for this pur- 
pose. Quantitative results are obtained with the tacky-dish tech- 
nique by counting the number of particles of each type that are pres- 
ent in a given area. It has been found that reflected illumination is 
most suitable. The field which is being counted can -be defined by a 
photographic grid fastened to the dish with tape, or by a hole of 
known size cut in black paper. A grid in the eyepiece of the micro- 
scope (Whipple disk) is also very convenient. The results are usually 
expressed as particles per square foot per hour, calculated from the 
area of the dish and the time of exposure. Low magnification (25 X) 
is used for general counting, while higher powers should be used for 
the examination of individual particles. 

In motion picture work, it has been found convenient to make a 
separate count of all particles having a maximum dimension of 0.002 
in. or greater, as it may be arbitrarily assumed that this size particle 
on a motion picture negative would produce a "window" or "star" on 



198 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

a print which would be objectionable on the screen, whereas anything 
smaller would* be noticed only in the form of "shower," where many 
such spots would appear at once on the screen. 

Many types of contamination are easily recognized when observed 
under adequate magnification, while other materials require further 
study before they can be positively identified. Simple chemical tests 
requiring only small amounts of reagents are the surest method of 
identification. Color reactions or spot tests may be used under the 
microscope. Fragments of emulsion can be detected rapidly by de- 
velopment and bleaching while film support is usually identifiable 
through its solubility in acetone or other solvents. When making 
chemical tests on various particles in the dishes it is advantageous to 
transfer the particle to be tested to a microscope slide. Only in 
special cases where a specific contaminant is sought is it advisable 
to add reagents to the entire dish. 

A tacky dish using reagent in the gel can be employed when metal 
dust is being studied. The reagents keep poorly, so the active mate- 
rial is made fresh each time. Extreme accuracy is not needed in . 
weighing the chemicals, but care should be exercised to avoid the 
presence of dust. The reagents can be prepared as follows : 

To one liter of 10 per cent gelatin solution, add 30 cc of water 
containing 

. 5 gm potassium f errocyanide 

. 5 gm potassium f erricyanide 
10 . cc cone, hydrochloric acid 

4 . gm calcium chloride 
50 . cc ethyl alcohol 

Finally, add 300 cc of glycerin to the mixture. Particles containing 
iron produce blue spots, while copper-bearing compounds give 
brown spots. 

Sedimentation data have been made more useful occasionally by 
keeping track of time. This is done by placing the dish upon a turn- 
table driven by a clock. An angular space representing about one 
hour is exposed by means of a hole cut in a stationery mask. Exami- 
nation of the chart will show how the sedimentation of dirt varies 
with time. There will be an uncertainty in the exact time, but this 
is usually unimportant since the information generally enables one 
to identify the cause by reason of prior knowledge of procedure. 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 199 

Results of a typical set of tacky-dish dirt count tests are given in 
Table 1 and shown graphically in Fig. 6. 

TABLE 1 

Dirt 

Count 

(No. of 

Particles 

PerSq .n. 

Sample No. Location of Test Plate Per Hr) 

la Negative developing machine, main air 16 

inlet duct. After air filters 
Ib Ditto, after cleaning and painting the 6 

main air inlet duct 
2a Negative developing machine drying 225 

cabinet, position 1 (directly in air 

stream after air has passed over dry- 
ing film) 
2b Ditto, after cleaning and painting 8 

louvers in drying cabinet 
2c Negative developing machine drying 7 

cabinet, position 2 (further removed 

from air inlet duct than position 1), 

after cleaning and painting louvers 

3 Printing room, on top of cabinet (8 ft 40 

from floor level) 

4 Cleaning room, on top of cabinet (8 ft 5 

from floor level) 

5 Table in assembly room on which nega- 20 

tive rolls are broken down into 
scenes. Test plate placed close to 
one end of rewind 

6 Splicing bench No. 1 in assembly room. 33 

Test plate placed within 8 in. of 
splicer 

7 Splicing bench No. 2, otherwise ditto 30 

8 Splicing bench No. 3, otherwise ditto 38 

It is apparent from examination of these data that an extremely 
bad condition existed in the developing machine drying cabinet at 
the start of these tests (sample No. 2a). Tests on the material picked 
up from the air stream by the tacky dish showed the presence of 
copper and a powdery material resembling pumice or talc. Investi- 
gation showed that the lower louvers under the drying cabinet con- 
tained a considerable amount of dusty material, with some greenish 
particles which analysis showed to be a copper compound. Further 
investigation brought to light the fact that the brass rollers in the 



200 



N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 



developing machine were periodically cleaned with a metal polish 
containing pumice and aqueous ammonia. This was wiped off the 
polished brass rollers, but a certain amount left in crevices, etc., had 
dried and fallen through the grating floor of the cabinet into the 
louvers below. 

As further proof that this dirt source was vitally important, small 
specks of foreign matter were removed from the processed negative 
film emulsion surface and analyzed. These showed the presence of 
copper and the same characteristic appearance under the microscope 
as the particles found in the tacky dish. 



230 




lo Ib 2a 2b 2c 3 45678 

SAMPLE NO. 
FIG. 6. Chart of dirt count, according to Table 1. 

After cleaning and painting the louvers and discontinuing the use 
of the pumice-base brass polish in favor of ordinary aqueous am- 
monia, a retest gave the result labelled "Sample No. 2b" in Table 1. 

In Table 2 and Fig. 7 are shown the results of tacky-dish tests made 
at another motion picture processing laboratory. These tests indi- 
cate the most dangerous conditions to be an excessive amount of lint 
found in the cleaning room, and an excessive amount of fibrous mate- 
rial, chiefly from the worker's gloves, in the daily assembly room. 
Steps were taken to correct these conditions by installation of suction 
equipment in the cleaning room, and by the use of laundering methods 
to render the cotton gloves more nearly lint-free. 



March, 1946 



MEASUREMENT AND CONTROL OF DIRT 



201 



Tape Samples. Dirt which has settled on surfaces can be 
observed in place if conditions permit, but is generally "lifted" if 
the location is in a darkroom. This procedure consists of pressing a 
piece of adhesive tape firmly against the surface of the area to be 
sampled, removing, and mounting for microscopic examination. 
Either black or white tape may be employed, depending on the color 
of the dirt picked up. Mounting is best done by using a cardboard 



& w 












85 
















80 


- 














75 


- 














70 


- 














65 


- 












. 


60 


- 


















55 


- 


















50 


- 










: -J:-:' 






' 


45 
















r^ 








40 


- 














^i 








35 


- 














^ 








1 


30 


- 














^ 










25 


i i 


















r^ 










20 


. 


::''.'; 








^y 








'-?;: 




K 










15 


- 










M 




''/-.' 




n 




p 









H 


























>^j-i 










10 


_ 


: : -'-'-' 








~~fs 




.''.-.' 




'/ * 












T^ 


5 




vi" 




:'-':- : .-': 

3S 




% 








~*\ 




1 




;:U- ;; 




r& 

*{&. 


^Y? 


cV 


^ 



10 



SAMPLE NO. 

FIG. 7. Chart of dirt count, according to Table 2. 



mask of distinctive color with a small cutout hole sandwiched between 
the sticky surface of the tape and a glass microscope slide. The mask 
prevents too tight contact between tape and glass slide, which would 
tend to make identification of the particles difficult. The method of 
lifting dirt offers a very rapid way of looking for rust or scaling paint 
and makes possible a rapid check upon housekeeping practice. A 
few examples are shown in Fig. 8. 

Microscopic Examination of Film. There are a number of 
observations that can readily be made with the aid of a microscope 



202 



N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 



TABLE 2 



Sample 
No. 



6 



S 



10 



Location of Test Plates 

Negative developing machine 
drying cabinet. Floor of 
4th bay 

Negative developing machine 
drying cabinet. Floor of 
4th bay 

Negative breakdown room, 
surface of work bench 



No. 
Particles 

Per 
Sq In. 
Per Hr 



Description of 
Test Plate 

Very clean. No fibers. 
Several pieces brown- 
ish material like rust 
or clay. No fine dust 

Same as No. 1 



24 



45 



63 



19 



29 



Few white fibers. Many 
short slivers of film 
base or emulsion. 
Much fine dust 

Blue wool fibers. Few 
large scrapings of film 
base. Much fine dust 

Many pink and white 
fibers. Few small 
opaque particles 

Few white fibers. Much 
fine dust 

Very few fibers, some 
colored. Some fine 
dust . Many film base 
and emulsion slivers 

Many w r hite fibers. 
Large green plastic- 
like particles. Some 
red fibers. Much fine 
dust. Some rust par- 
ticles 

Very few fibers. Few 
opaque particles, 

brown or green in 
color. Little fine dust 

Many white fibers. 
Some film base scrap- 
ings. Some fine dust 

Note: The particle counts refer only to particles larger than 0.002 in. in any 
one dimension. The term "fine dust" refers to those particles too small to be 
counted. 

that will serve to elucidate screen examination of the finished print 
and limit the possible sources of film contamination. Some of these 
may be listed : 



Negative breakdown room. 
Back of splicer 

Negative cleaning room. 
Work bench, center, rear 

Printing room. Top of rheo- 
stat box for No. 11 printer 

Printing room. Floor, be- 
tween printers No. 4 and 
No. 5 

Daily assembly room. Work 
bench surface, center east 
side 



Negative cutting room. Work 
bench surface, S. E. corner 



Negative cutting room. Back 
of splicer 



17 



36 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 



203 



Rusty 
Spring 



Wood, 
Cinders, | 
etc. 




Brass" 
Grind- 
ings 



Door 
Latch 



FIG. 8. Tape samples showing "lifted" dirt. 



204 N. L. SIMMONS AND A. C. ROBERTSON Vol 46, No. 3 

(1 ) Position of the particle on the film . 

(a) Emulsion or support side. 

(b) Concentrated in the neighborhood of splices. 

(c) Consistently near the starts or ends of rolls. 

(2) Confirmation of actual particle rather than a "shadow pattern." 

(a) Often the "windows" or "stars" observed on prints are shadow pat- 
terns of actual dirt particles adhering to the negative or of loose dirt 
lying between the negative and positive films during the printing 
operation. 
(5) Photographic effect. 

(a) Sensitized or desensitized area (sometimes both) surrounding the par- 
ticle after development. Many substances, notably metals, will 
react with the photographic emulsion and produce characteristic 
spots. The area of the spot may be many times the size of the actual 
particle. 6 

(4) Distance between particles. 

(a) Repeat pattern, owing to contamination on a sprocket or idler roller. 

(5) Chemical nature of the particle. 

(a) Solubility tests. 

(b) Color reactions of metals. 

When the processed film contains an objectionable amount of dirt, 
this is proof that an abnormal condition existed at some point in its 
handling by the user, or in some stage of manufacture. The amount 
of contamination on or in the emulsion of modern motion picture 
films before exposure is extremely low. If any large amount is sus- 
pected on the film as received, a strip should be taken directly from 
the can without excessive handling and examined in white light. 

The high standards of quality of present-day motion pictures makes 
it well-nigh imperative that some measure of dirt control be instituted 
as an integral part of the production system, providing a safeguard 
against contamination of the finished product. 

The authors wish to express their appreciation to Emery Huse, 
under whose guidance the motion picture laboratory investigations 
herein described were undertaken, and to the management of the 
several laboratories where such tests were carried out. 

We also want to express our thanks to colleagues at Kodak Park, 
particularly J. M. Mulder, F. G. Van Saun, and J. R. Peer, who aided 
in the development of many of these techniques. 

REFERENCES 

1 ROBERTSON, A. C., MULDER, J. G., AND VAN SAUN, F. G.: "Measuring 
Smokes and Rating Efficiencies of Industrial Air Filters," Ind. Eng. Chem. (Anal. 
Ed.), 13 (1941), p. 331. 



March, 1946 MEASUREMENT AND CONTROL OF DlRT 205 

2 CRABTREE, J- 1., MATTHEWS, ,G. E., AND MUEHLER, L. E.: "Materials for 
the Construction of Photographic Processing Apparatus," Kodak Research 
Laboratories, Eastman Kodak Co. (Rochester, N. Y.), 1945. 

3 HENN, R. W., AND CRABTREE, J. I. : "Calcium Scums and Sludges in Photog- 
raphy," J. Soc. Mot. Pict. Eng., 43, 6 (Dec., 1944), p. 426. 

4 BLOOMFIELD, J. J., AND DALLAVALLE, J. M.: "The Determination and 
Control of Industrial Dust," Pub. Health Bull. No. 217, Government Printing 
Office (Washington, D. C.), 1935. 

6 DRINKER, P., AND HATCH, T.: "Industrial Dust," McGraw-Hill Book Co. 
(New York), 1936. 

6 CRABTREE, J. I.: "Stains on Negatives and Prints," Eastman Kodak Com- 
pany (Rochester, N. Y.), 1945. 

DISCUSSION 

QUESTION: How do you launder your special uniforms? Can commercial 
laundries do the work? 

DR. ROBERTSON: A great deal of our laundry work is done by a commercial 
laundry. We have had good results, which may be due in part to close technical 
co-operation. 

QUESTION: What about oil on the film? There has been no mention made 
of the presence of oil in the compressed air used for the squeegees. 

DR. ROBERTSON: You are quite right about the trouble caused by the pres- 
ence of oil on film. The best air compressors for this purpose are the- ones which 
are water-sealed. 

QUESTION: What is the name of the compound used to prevent the growth of 
slime in wash-water tanks? 

DR. ROBERTSON: Sodium pentachlorphenate is the chemical name of the 
material. It is sold under various trade names, such as Dowicide, Nalco, and 
Santobrite, and the manufacturers' salesmen will be glad to help you further. 
Since the material is not very soluble in water, it is sometimes added to the 
water in the form of a stock solution. Sodium pentachlorphenate has been used 
by a number of West Coast laboratories, but I do not know if it is employed in 
New York City. It can also be used in air washers, but it is best to use it in 
the pure form. If sodium carbonate is present, some foaming may ensue. 

QUESTION: What about radium buttons? The material tends to powder and 
give trouble. 

DR. ROBERTSON: They should be hermetically sealed, as in glass or plastic, 
and handled only by authorized personnel. We have also had trouble with girls 
wearing "black-out" jewelry. There are other troublesome materials such as 
face powder, etc. I could go into great lengths about similar necessary pre- 
cautions, but not everyone is interested in that matter. 



A NEW RECORDER FOR 16-MM BUZZ TRACK* 
M. G. TOWNSLEY** 

Summary. A new recorder is described which produces direct recorded 16-mm 
buzz track film of exceptional accuracy. The track has a dense center portion to re- 
duce ground noise and the edge modulations form a continuous exposure with the 
center portion. 

In a paper published in 1938, Kellogg 1 pointed out the advantages 
of recording buzz track film directly rather than making it by printing 
from a negative, and described a recorder for producing directly re- 
corded film. This recorder used a single "chopper" wheel to produce 
edge modulations having frequencies of 300 and 1000 cycles on op- 
posite edges of the track. The chief disadvantage of the film so -pro- 
duced was that the unmodulated portion of the track was clear film 
and had a tendency to produce considerable ground noise. 

After some experience with a recorder similar in principle to 
Kellogg's, it was found that the ground noise could be very appre- 
ciably reduced by separately printing an exposure in the unmodu- 
lated area. This was done by removing the modulating chopper 
wheel, placing a mask to define the edges of the strip to be printed, 
and re-exposing the film on which the modulation had already been 
exposed. This left a narrow line of light between the outer edge of the 
center strip and the inner edge of the modulation, which tended to 
fog slightly and reduce the effective track width. 

A track was finally evolved in which the center exposure was ac- 
curately controlled for location and the width, and the edge modula- 
tions were printed extending into the center strip for a few thou- 
sandths of an inch on each side, eliminating the white line. This pro- 
duced some modulation near the edge of the center strip but the 
density could be kept sufficiently high so that the output from this 
slight modulation was well within the tolerance limits. 

* Presented October 15, 1945, at the Technical Conference in New York. 
* * Bell and Howell Company, Chicago, 111. 

206 



A NEW RECORDER FOR 16-Mn Buzz TRACK 



207 



When the American Standards Association Committee on Photog- 
raphy and Cinematography Z52 adopted the War Standard for 16- 
mm buzz track film, 2 the Standard specified a track having the 
narrow white line between the outer edge of the center strip and the 
inner edge of the modulation. 

The necessity for producing considerable quantities of buzz track 
for our own use and for the Signal Corps pointed to the desirability 
of producing a new recorder, which would replace the nearly worn out 
unit then in use and, at the same time, improve the accuracy and 
convenience of producing the 
track. It is the purpose of this 
paper to describe the recorder for 
the record and point out some 
of the desirable features of buzz 
track produced with it. 

It seemed desirable to continue 
the feature of making the edge 
of the center strip the accurate 
element, with the modulation ex- 
tending from it. This considera- 
tion, and the necessity for elimi- 
nating any modulation from the 
edge of the center strip, dictated 
the making of the entire track in 
a single exposure, with the modu- 
lation produced by interrupting 
the light along both edges of the 
track. This is accomplished by 
the use of a long slit which is imaged on the film and interrupted at 
the ends by 2 separate chopper wheels. 

Fig. 1 shows a short section of buzz track produced in this manner. 
The basic dimensions conform to ASA War Standard Z52.10-1944, 
but, as described, the track differs from the Standard in omitting the 
narrow light line just inside the end of the modulation, and in having 
the width of the black, continuous center strip equal to the standard 
width of the scanned area. The Signal Corps has accepted large 
quantities of this film as being in accordance with the American War 
Standard. 

The recorder itself is simply but ruggedly constructed, and is made 
as convenient as possible to adjust and operate. All elements are 




FIG. 1. Buzz track. 



208 



M. G. TOWNSLEY 



Vol 46, No. 3 



mounted on a main frame which consists primarily of 2 steel plates at 
right angles to each other. Fig. 2 shows a general view of the complete 
recorder, and Fig. 3 shows the optical arrangement. The lamp house 
is at the left end, and contains a 10- v, 7V 2 -amp exciter lamp arranged 
to be readily adjustable for proper illumination of the slit. A con- 
denser LI close to the lamp images the filament a few inches to the 
right of the slit and a field lens L 2 just ahead of the slit images the fila- 
ment in the objective lens L 3 to provide uniform coverage of the wide 
track required. 




FIG. 2. Buzz track recorder. 

The slit is vertical and is set to a width of approximately 15 mils. 
Since the objective is set to a magnification of 0.1, this gives a beam 
width at the film of 1.5 mils. Just to the right of the slit are the 2 
chopper wheels with their independent motors. Separate motors 
were used because there were available a quantity of small governor 
controlled motors with appropriate speed range. The chopper 
wheels have 3 and 10 notches, respectively, in their edges and ac- 
curately running outside diameters. It is the outside rim of these 
choppers which is imaged on the film to give the accurate track loca- 
tion and width. The notches in the edge of the chopper wheels 
allow light from the ends of the slit to reach the film and produce 



March, 1946 A NEW RECORDER FOR 16-MM BUZZ TRACK 209 

the modulation. Motor speed and number of notches were chosen 
in relation to the translational speed of the film to give frequencies 
of 1000 and 300 cycles per sec on the finished film. 

Each motor is mounted on a separate angle plate as shown in Fig. 4, 
which is a detail of the chopper and slit assembly with the light-tight 
cover removed. The angle plates are mounted in slides on the main 
frame so that they may be moved up or down as required for adjust- 
ment of the location of the track edges. Separate micrometer heads 
are provided for each motor so that adjustment of the location of the 
edges of the track to within 0.1 mil is quickly and easily accomplished. 

The film is transported horizontally from a feed flange through 
sprockets and an idler roller system, borrowed from a sound pro- 



h 






FIG. 3. Recorder optical system. 

jector, to a drum on which the recording is done, and then to a take-up 
flange. Holdback and take-up tension is provided by friction wash- 
ers which support the flanges. This method of handling film with the 
roll laid flat is very convenient in devices of this type, keeping the roll 
in good condition, permitting the use of laboratory packed film with- 
out elaborate guide flanges, and keeping the film above all lubricated 
bearings and thus free from oil. 

Edge guiding at the exposure point is provided by a stationary 
block which is surfaced with polished hard chromium plate. The 
sound track edge of the film is held against this guide by a slight 
crown on the film drum analogous to the crown on a flat pulley which 
is used to keep the belt running on the center of the pulley, and by a 
light spring which bears on the opposite edge of the film. The drum is 
supported in ball bearings and is driven by the film. It was found 



210 M. G. TOWNSLEY Vol 46, No. 3 

that the use of a fixed guide gave better precision with a ball-bearing 
drum than a flange on the drum because of minute roughness in the 
bearings available. 

The sprockets and sprocket-drive mechanism are driven through 
a fiber idler gear from a synchronous motor. A separate take-up 




FIG. 4. Chopper and slit assembly detail. 

motor is provided to drive the flange on which the take-up flange 
rests. While this arrangement results in the use of a total of 4 motors 
on the recorder, it was simpler to use multiple motors than to make 
the necessary mechanical devices to drive everything from one motor. 
Electrical controls are provided in a separate case connected to the 
recorder by a 6- wire cable. Current for the lamp is provided by a 
constant voltage transformer, step-down transformer, and full- wave 
rectifier, and is controlled by a rheostat and ammeter. Motor cur- 



March, 1946 A NEW RECORDER FOR 16-MM Buzz TRACK 211 

rent for the various motors is provided, and controlled by separate 
switches. 

This recorder produces buzz track of minimum ground noise and 
excellent accuracy in a single pass through the recorder. It has been 
found possible to consistently maintain an accuracy of track location 
well within the limits specified in American War Standard Z 52.10- 
1944. 

REFERENCES 

1 KELLOGG, E. W.: "A Recorder for Making Buzz Track," /. Soc. Mot. Pict. 
Eng., XXX, 2 (Feb., 1938), p. 150. 

2 AMERICAN STANDARDS ASSOCIATION, NEW YORK: War Standard Z52.10- 
1944. 



NEW PERMANENT MAGNET PUBLIC ADDRESS 
LOUDSPEAKER* 



JAMES B. LANSING** 



Summary. This paper describes the permanent magnet type of Duplex loud- 
speaker and its use in theater public address systems. Its efficiency and frequency 
response is shown. Special types of baffles for the sides of the stage and above the 
screen are described. Because of its high efficiency and wide angle of distribution and 
extreme frequency range along with its small physical size, it provides a 2-way horn 
system with many advantages over those in present use. 



This paper describes the new Duplex loudspeaker with permanent 
magnet fields, and its application to theater public address systems. 
At the Hollywood Technical Conference in October, 1943, an earlier 
model was described 1 which used separately excited field structures. 
The general performance and operating efficiency of the new speaker 
has been improved through the use of new manufacturing techniques 
and new materials. Fig. 1 shows the 604 Duplex loudspeaker. 

Although permanent magnet speakers have been manufactured for 
years, they have been for the most part of very low efficiency because 
of the poor operating characteristics of the magnetic materials which 
have required large, costly, structures for the efficient use of the energy 
available from them. As a consequence, where attempts have been 
made to obtain high loudspeaker efficiency, it has been the practice to 
use the older magnet materials in an inefficient manner, usually in the 
form of ring castings, with the resulting very high stray fields with 
which most engineers are familiar. 

The use of the Alnico No. 5 magnets has contributed considerably 
to the high performance and efficiency of the new Duplex loudspeaker. 
Through the use of these new highly efficient magnets, which are 3 
times as powerful as any magnets previously used, it has been possible 
to obtain the high flux densities and the large total flux required for 
high-efficiency loudspeaker operation, which has been obtainable in 

* Presented May 14, 1945, at the Technical Conference in Hollywood. 
* * Altec Lansing Corporation, Hollywood. 
212 



NEW MAGNET PUBLIC ADDRESS LOUDSPEAKER 213 

the past only with electromagnets. The new magnets are used in this 
design in the same manner as field coils ; that is, they are surrounded 
by the field pot or the return circuit. The stray flux or external field 
surrounding these field structures is lower than that generally en- 
countered in the electromagnetic speakers, even though the flux in the 
gap is greater than has ever been used before in commercial practice. 
The stray fields are low enough so that these speakers may be used 
in close proximity with oscilloscopes without affecting their operation. 




FIG. 1. Altec Lansing 604 Duplex loudspeaker. 

The use of permanent magnets permits the voice coils to operate at 
considerably lower temperatures than would be the 'case if the sur- 
rounding magnetic structures were heated by the power dissipated 
by the field coils of the electromagnetic structures. 

The high-frequency diaphragm of the Duplex loudspeaker utilizes 
tangential corrugations in the compliant portion surrounding the 
center dome (see Fig. 2). The increased freedom of movement of 
this new diaphragm over the older types which use annular com- 
pliance permits it to handle greater excursions without transmitting 
undue flexure stresses to the center dome, thereby increasing the peak 
amplitudes which it will handle and increasing its life. These new dia- 



214 



J. B. LANSING 



Vol 46, No. 3 




phragms are easily replaceable. Accurately positioned dowel pins 
in the top plate and corresponding holes in the rings of the diaphragm 
assembly assure proper alignment of the voice coil in the gap. 

Another contributing factor to the better performance of the new 

speaker is the use of edgewise 
wound ribbon wire in the voice 
coil. High baking temperature 
varnish is applied to the ribbon 
as it is being wound. The 
winding is clamped securely so 
that the turns are firmly pressed 
together and baked at tempera- 
tures considerably in excess of 
those encountered in the opera- 
tion of the speaker. The re- 
sultant coil, therefore, is quite 
stable and is not affected by 
the peak currents which it must 
handle in normal operation. 

The ribbon wire permits the 
use of 27 per cent more con- 
ductor in the same space as 
compared with round wire, and 
thus provides the maximum 
utilization of space available 
in the gap. Since these rib- 
bon-wound coils are always 
single layer and the voltage 
developed between turns is very 
small, insulation requirements 
are minimized and the heat 
radiation of the coil is im- 
proved. 

The high-frequency response 
of the Duplex loudspeaker has been improved through the use of 
the ribbon coil construction by increasing the sensitivity of the high- 
frequency speaker without increasing the mass of the moving 
system. 

The low-frequency response has been improved by the use of in- 
creased magnetic flux, the ribbon coil construction, and the use of a 




FIG. 2. High-frequency diaphragm 
and voice coil assembly for 604 Du- 
plex loudspeaker. 



March, 1946 NEW MAGNET PUBLIC ADDRESS LOUDSPEAKER 



215 



larger diameter voice coil to provide better coupling between the cone 
and the forces acting upon the coil (see Fig. 3). The increased coil 
diameter also decreases the depth of the cone and increases its effective 
stiffness thereby causing it to act more nearly as a piston. The over- 
all gain in performance with 
the Duplex loudspeaker has 
been such that satisfactory re- 
sults are obtained at a cross- 
over frequency of 2000 cps. 
The crossover network is shown 
in Fig. 4. 

Permanent alignment of the 
pole pieces and gap are assured 
by the use of centering rings, 
fitting snugly over the inner 
pole pieces and into recesses in 
the top plates. A heavy die- 
cast cone housing is used in 
order to insure proper align- 
ment of the low-frequency 
voice coil. The multicellular 
horn is a single bakelite cast- 
ing with a mounting base at- 
tached by means of drive screws. 

The application of public 
address and speech reinforce- 
ment systems to the theater 
has posed many problems. 
Such systems in the past have 
consisted of the microphone 
and amplifier channels and a 
multiplicity of cone speakers 
distributed about the theater. 

In the applications where cone speakers have been used, the over-all 
efficiency of the system has usually been very low and the quality very 
poor. The polar distribution patterns of cone speakers has prevented 
uniform sound distribution throughout the auditorium and the 
marked resonance peaks of the cones at various frequencies have 
made it impossible to obtain any appreciable reinforcement without 
serious feedback or "howls." Such systems have proved as unsatis- 




FIG. 3. Low-frequency diaphragm for 
604 loudspeaker. 



216 



J. B. LANSING 



Vol 46, No. 3 



factory for theater public address work as they have for use with 

motion pictures, since no amount 
of electrical compensation could 
make up for their inherent defi- 
ciencies. Microphone character- 
istics or amplifier capacity had 
little or no effect upon the final 
result. 

The use of conventional expo- 
nential horns either in cluster 
or in multicellular array has 
provided improved distribution 
patterns and greater sound re- 
inforcement. Although the 
quality is greatly improved over 
that obtained with systems 
using cone speakers alone, the 
limitations of the speakers that 

FIG. 4. N-2000-A dividing network. have been available hereto- 
fore for this type of service 

have been a serious drawback. Usually, too, the low-frequency 
cutoff of these horns is too high to permit satisfactory reproduction 





FIG. 5. 613 loudspeaker. 



March, 1946 NEW MAGNET PUBLIC ADDRESS LOUDSPEAKER 



217 



below 300 cps. Where cone speakers on a flat baffle have -been used 
with these systems to improve the low-frequency response, the over- 
all efficiency level of these systems has necessarily been reduced to 
that of the cone speakers 
used. The large size of such 
speaker systems makes it diffi- 
cult, or impossible, in most 
cases to find the space in the 
stage area where they will be 
out of the way, or will not 
present an unsightly appear- 
ance in the auditorium. 

The Duplex loudspeaker dis- 
cussed earlier in this paper will 
prove invaluable for the fore- 
going applications. Also, it 




FIG. 6. 606 loudspeaker. 



can be used advantageously as 
a monitoring speaker in those 
cases where the high quality of the pickup is to be maintained for re- 
cording or broadcasting purposes. 

This loudspeaker can be mounted in an enclosure as small as 4 cu ft 




FIG. 7. 612 loudspeaker. 



218 J. B. LANSING Vol 46, No. 3 

for small auditoriums. Where greater coverage is required for larger 
auditoriums, 2 or more speakers can be enclosed, as shown in Fig. 5. 
With this type of enclosure, maximum illusion of stage presence 
is obtained. Because of its small size it can readily be placed on stage 
near the proscenium arch without being too conspicuous. 

Odd-shaped cabinets (see Fig. 6) can be made as required to meet 
particular adaptations so long as they are deep enough to house the 
speaker, which is 12 in. from front to back. 




FIG. 8. 614 loudspeaker. 

The 612 loudspeaker for use in theaters and industrial applications 
is shown in Fig. 7. The cabinet is finished in dull gray and its volume 
is6Y 2 cuft. 

The 614 loudspeaker (see Fig. 8) for use in portable public address 
systems utilizes a cabinet of only 4Y 2 cu ft, finished in dull gray. 

The uniform frequency response and wide distribution angle of the 
Duplex loudspeaker make it possible to obtain reinforcement levels 
above those which can be obtained with other systems for the same 
auditorium positions. 

Amplifier power capacity requirements when using these speakers 



March, 1946 NEW MAGNET PUBLIC ADDRESS LOUDSPEAKER 219 

are approximately equal to those required for the most efficient back- 
stage speakers. A maximum of 30 w of amplifier capacity should be 
installed for each Duplex speaker in the system. 



REFERENCE 



1 LANSING, J. B.r "The Duplex Loudspeaker," /. Soc. Mot. Pict. Eng., 43, 3 
(Sept., 1944), p. 168. 








SPECIALIZED PHOTOGRAPHY APPLIED TO ENGINEERING 
IN THE ARMY AIR FORCES* 

P. M. THOMAS** AND C. H. COLESf 

Summary. High-speed motion pictures, high-speed still pictures, special re- 
cording devices, and color photography have all played important parts in the design 
and engineering program of the Army Air Forces. This paper outlines some of the 
more important applications. 
\ 

Historical. Although sporadic attempts hard been made from 
time to time at Wright Field to employ the latest methods of photog- 
raphy toward the solution of aircraft problems, no concentrated 
effort was made to press its advantages until the Photographic En- 
gineering Branch of the Technical Data Laboratory, Engineering 
Division, was organized in the middle of 1943. Now it has grown to 
a total personnel of 65, and has spread into 2 buildings. Its work 
has been deemed so important that its program has remained vir- 
tually unchanged since V-J day. 

Organization. The organization gives a good idea of how a 
problem is attacked. When a laboratory on Wright Field requests 
work to be done, the Projects Branch surveys the problem. The 
varied experience of its project engineers is of tremendous value in 
deciding a method of attack. One of these engineers makes an out- 
line of the essential information to be obtained. For problems in- 
volving special electronic flash equipment, Dr. H. E. Edgerton and 
Gjon'Mili, staff consultants for this organization, may be called in 
for advice. 

As an example, the Jet Powered Unit requested that velocity and 
acceleration studies be made of the take-offs of the newly completed 
American version of the German V-l flying bomb. A project en- 
gineer flew to the test base where the launching ramps were under 
construction. He decided that a camera tower had to be constructed 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Major, fCaptain, AAF, Special Photographic Services Section, Wright 
Field, Ohio. 
220 



HIGH-SPEED PHOTOGRAPHY 



221 



at a certain location and 10-foot distance markers erected along the 
launching track. These were built under his direction. 

The project was now turned over to the Field Branch and a crew 
was sent with high-speed motion picture cameras, batteries, timers, 
and developing equipment. The first few take-offs of the experi- 
mental bombs were filmed and the records flown back to Wright 
Field, where the Analysis Branch took over. Here the films were 
studied and velocity and acceleration curves were drawn from the 
information obtained from the films. 

The curves and tables were now returned to the same project en- 
gineer who initiated the action. From these data he assembled the 




FIG. 1. 



Take-off of jet propelled bomb as photographed by automatic camera 
at Proving Ground. 



final report and submitted it to the engineers of the Jet Powered 
Unit. This completed the job. 

An interesting side light on this particular project was that the 
films showed not only that the first flying bombs failed to attain ade- 
quate speed to enable them to fly, but also the exact cause of the 
failure. It was poor rocket placement and consequent loss of power 
during take-off. 

.The high-speed photographic equipment does only a part of the 
work of the organization. Special photographic triangulation 
methods are employed to determine the height of aircraft and bombs 
at any point, the path of an airplane or falling projectile, the suc- 
cessive positions in space of a helicopter or a parachute. 

When existing cameras are inadequate for a required purpose, the 



222 P. M. THOMAS AND C. H. COLES Vol 46, No. 3 

Installation and Fabrication Unit designs its own or turns its speci- 
fications over to a commercial company. A specially constructed 
camera of this type is being used to photograph the indications on 
instrument panels during flight tests or wind-tunnel tests. One of 
the first to come off the production line was used to make the famous 
pictures of the explosion of the first atomic bomb in New Mexico. 

Color photography has grown in importance as a recording me- 
dium for engineering data. Corrosion, combustion, color signals, 
moisture detectors, medical subjects, all require color photography 
for adequate recording. Complex production graphs and engineer- 
ing charts also require color for clarity. To make the color records 
of value, they must be capable of being printed for reports, so an ex- 
tensive color printing service has been instituted which can turn 
out as many as 500 prints per week. The original transparencies 
are processed in our own laboratory.' The color prints are made by 
direct contact, or by enlargement from the original transparencies 
and are also entirely processed with our extensive facilities. 

The organization is constantly experimenting with new methods 
and applications for enlarging its scope and attacking new problems. 
New applications of infrared, x-rays, photomicrography, radar, 
stereoscopic processes, cathode-ray oscilloscopes are constantly being 
tried to discover new approaches for the solution of problems pre- 
sented to this organization. The limits of photography as an aid to 
engineering have by no means been reached. 

High-Speed Photography. High-speed photography as prac- 
ticed at Wright Field may be conveniently divided into 2 general 
classifications: high-speed motion pictures, high-speed still pictures. 
Each of these may be subdivided into qualitative records and quan- 
titative studies. 

High-Speed Motion Pictures Continuous Light. To make the 
high-speed motion pictures, several pieces of equipment are in use. 
The Western Electric Fastax camera, in both the 8-mm and 16-mm 
sizes, is used for the bulk of the projects. Although the Eastman 
Type III 16-mm camera is gaining favor, each camera has certain 
advantages for various types of problems. 

These cameras operate on the principle of continuous film motion, 
the individual frames being defined by a glass prism which, by rotat- 
ing, moves the image formed by the camera lens along with the ad- 
vancing film. 

The definition obtained with this type of optical compensation is 



March, 1946 HlGH-SPEED PHOTOGRAPHY 223 

not as good as with the standard intermittent motion, but it is ade- 
quate for any but the smallest detail. The best use of these cameras 
is made by filming close-ups of the most important action and in this 
way not depending upon rendition of small detail. 

At full speed, the 100 ft of film take about l l / z to 2 sec to pass 
through, so exact timing of the starting of the cameras is essential. 
It is all too easy to have the important action occur after the film has 
passed through the camera. 

In one application where it was desired to record the final velocity 
of a cart falling in a vertical track, a switch was installed on the track 
to turn the camera on just before the cart appeared in the picture, 
thus assuring that the camera would be running at that time. The 
switch was then moved up the track until the camera started x /2 sec 
before the cart arrived in the scene. In this way the camera was 
allowed to come up to its full speed of 4000 pictures a sec before the 
cart appeared. 

To measure the velocity of the cart it was necessary to determine 
the distance traveled in a certain time The track was marked in 
one-inch distances by painting the track white and laying on strips 
of black scotch tape one inch wide every other inch. To improve the 
accuracy of the measurement a vernier scale was painted on the cart 
so it would move along the track markings. Thus a reading every 
tenth of an inch could be made. 

The time record on the film was made by means of a 200-cps spark 
originating in a vibrating-reed timer built for the purpose by our 
organization. A vibrating reed has its output voltage stepped up by 
a transformer to a point where it trips the grid of a Strobotron tube. 
The output of the Strobotron is put through a spark coil on the camera 
which is connected to the sparking electrode in the camera body. A 
spark flashes against the film sprocket, the light of the discharge 
making a small fog mark on the edge of the film every 5 milliseconds. 

Thus time a"nd distance were recorded on the same film from which 
velocity may be calculated. By drawing the curve of velocity 
against time it is possible to take the slope of the curve at any point 
and so calculate acceleration. 

In this way a complete record is obtained not only of velocity and 
acceleration but of the appearances of the action in slow motion for 
visual study. 

Lighting of the subjects to* the required high intensity is provided 
by R-2 photofloods of small objects and up to 10,000-w units for 



224 P. M. THOMAS AND C. H. COLES Vol 46, No. 3 

larger areas. In bright sunlight, a maximum of 2000 frames per sec 
is all that the light will permit and still produce adequate exposure 
from light-colored subjects. For higher camera speeds, the subject 
must be illuminated with additional lighting units. A light truck 
equipped with four 3200-w floodlighting units and provided with its 
own generator was found to be a useful item for work in the field. 

Processing. Because the exposure per frame of films made with 
the Fastax or Eastman Type III cameras is of the order of Vsooo 
sec or less, the films are normally on the underexposed side. Special 
processing to obtain a printable film density is usually required. A 
fresh D-76 formula gives good density provided the development is 
carried from 30 to 60 min. To reduce the time of development, a 
more energetic print type of developer will cut the time to about 7 to 
10 min. With this stronger developer it is necessary to use an anti- 
foggant to hold down the background fog; 6-nitrobenzimidazole has 
been found very effective for this purpose. 

Removal of the opaque backing from the film has been found rela- 
tively easy with Eastman Kodak Super XX by a simple squeegeeing 
with a viscose sponge during final washing. The film must be doubly 
perforated similarly to "double eight," as usually supplied for 8-mm 
cameras. It is spooled in 100-ft lengths and is a reversible-type film 
although we develop it as a negative for convenience and extra speed 

In the laboratory, rack and tank development has proved more 
practical than machine processing because of the long developing 
time required to obtain sufficient image density. In the field a G-3 
tank is capable of producing good results. The roller in this tank is 
replaced with a viscose sponge to aid in removing the backing from 
the film. A collapsible drying rack designed for the Air Corps is a 
convenient accessory to hold the film while the moisture is evaporat- 
ing from it. 

Analysis of the Film. After processing, the film must be studied 
for information that will enable performance curves to be drawn for 
the subject under analysis. The timing marks along the edge of the 
film are usually reduced to a frames-per-second figure for important 
parts of the film where the action occurs. Where the whole film is to 
be analyzed, a curve is drawn of film length measured in feet from the 
beginning of the film against frames per second. In this way, the 
time interval measured from one frame to the next may be taken off 
the curve for any part of the reel while it is measured on a footage 
counter. 



March, 1946 HlGH-SPEED PHOTOGRAPHY 225 

The action is viewed frame by frame with special projectors origi- 
nally designed to analyze gun sight aiming point camera records. 
The image may be projected on a screen up to exactly original size 
and measurements of distance thus made directly on the screen. By 
interposing a mirror in the beam of the projector and reflecting the 
image back toward a translucent screen near the projector, the analyst 
may operate the projector and measure the screen without leaving his 
chair. A scale on the floor along which to slide the mirror enables 
the operator to consult a table and so enlarge the image to any de- 
sired extent without trial and error by setting the mirror at prede- 
termined distances from the screen. 

Large transparent protractors and scales to use on the screen en- 
able the analyst to work quickly and accurately. After the points 
for the velocity curve have been plotted, a special tangent scale de- 
vised by our chief analyst is used to obtain the points to plot the 
curve of acceleration. 

Achievements with the High-Speed Cameras. While the films 
made with high-speed motion picture cameras often appear spec- 
tacular, it is usually the more prosaic looking picture that produces 
the most significant results. The close-up of a wheel on the landing 
gear of a B-24 Curing the process of making contact with the ground 
during an actual landing is very dull screen fare, but it yielded curves 
and figures that explained a great deal about the flexures a tire under- 
goes during the violent impact at landing. 

Studies have been made of aircraft machine gun malfunctioning 
which proved the correctness of the theory of one of Wright Field's 
experts and revised the thinking of the gun manufacturers. 

Under the analytical eye of the high-speed camera have come aerial 
camera shutters, jet propulsion engines, bursting propellers, exploding 
oxygen containers, explosively operated radio antennae, electrical 
relay actions, manual gun charging operations, aircraft launching de- 
vices, and a host of other engineering projects. 

Intermittent Light High-Speed Motion Pictures. Another im- 
portant piece of equipment in use is the Edgerton flashing light high- 
speed camera. As is well known to most engineers in this field, this 
camera utilizes special gaseous discharge lamps whose flash is so short 
that it stops the action not only of the subject but of the continuous 
moving film in the camera as well. The film is 35-mm in width and 
100 ft long. The film passes through the camera in I 1 /* sec when the 
driving motor is set to full speed, taking 1500 pictures per sec. A 



226 P. M. THOMAS AND C. H. COLES Vol 46, No. 3 

contactor on the main sprocket wheel fires the lamps every time a 
new frame is in position back of the aperture plate. 

A spark electrode in the Edgerton camera places a time record on 
the film so that time duration, velocity, and acceleration may be 
measured. 

Incidentally, a comparison of the 3 types of high-speed cameras 
the Fastax, the Eastman, and the Edgerton brings to light the fact 
that 100 ft of film passes through each in 1V 2 sec at full speed. The 
linear film velocity is the same, therefore, in each camera, the dif- 
ferent frames per second rates being a result of the difference in frame 
size. 

The Edgerton camera can be used only in subdued light because 
the lens is open all the time; therefore its operation is restricted to 
laboratory applications. The shortness of the flash, which amounts 
to Viooiooo sec, and the relatively large frame size compared to the 
8- or 16-mm films made by the other cameras make possible the re- 
cording of greater detail in the pictures. Single frame enlargements 
up to 8 X 10 in. of the important phases of the action are readily 
made for inclusion in reports, a valuable aid in explaining data. 

Because of the shortness of the flash, normal speed films are barely 
exposed. Even the fastest films leave a great deal to be desired be- 
cause the severe reciprocity failure of the film reduces the effective 
exposure considerably. The blue color of the discharge lamp's flash 
utilizes only a portion of the wide spectral sensitivity of the fast pan- 
chromatic emulsion. 

Adding all these restrictions together pointed to the need for finding 
a high-speed blue-sensitive emulsion that could be developed vigor- 
ously. A blue-sensitive film made especially for recording the fluoro- 
scopic screen of x-ray apparatus was finally adopted as incorporating 
all the features desired. This film still has to be developed for 30 to 
60 min but the results are reasonably satisfactory. 

The coolness of the flashing light technique, as contrasted to the 
incandescent glare of the continuous light camera, indicates that for 
biological pictures and subjects whose actions would be affected by 
temperature rise, the Edgerton equipment would prove superior 
It can, however, photograph only relatively small objects because of 
the low light output. 

The continuous light cameras are lighter, smaller, cheaper, and 
simpler to operate and will make pictures in daylight. Each type 
of apparatus has its particular advantage and application. 



March, 1946 HlGH-SPEED PHOTOGRAPHY 227 

Sequence Flashing. Some actions are too fast for even the 
high-speed cameras to catch. What is more, the bursting of an air- 
plane propeller under increasing speeds cannot be anticipated, so it is 
impossible to start a motion picture camera in time to be operating at 
full speed at the exact moment required. Even if the camera were 
operating at the proper time, it is doubtful if more than 2 frames would 
record the action. A different technique had to be evolved to handle 
problems of this nature. 

If a series of electrical discharge lamps were lined up and their 
condensers charged, they may be fired in sequence at almost any 
rapidity desired. For relatively small ' but fast subjects such as 
bullets, 6 Edgerton micron 1 ash units were assembled. These units 
emit a flash of light whose duration is 1 /m,m sec. The condensers 
in each unit require several seconds to charge from a 7000-v supply, 
but once charged the units may be fired one after the other in rapid 
succession. 

A sequencing device was designed and constructed that would fire 
each lamp in turn electronically from x /i2 sec to 1 /20,ooo sec between 
flashes. By means of a microphone feeding into an amplifier, the 
sound of the gun initiates the sequence and the lights flash in suc- 
cession. Of course, the picture is made in darkness, the flashes of 
light exposing the film. The picture is taken on an ordinary still 
camera loaded with fast blue-sensitive film which is developed 
vigorously. In this way pictures of bullets may be photographed 
striking armor plate and shedding their jackets. A series of bullet 
images appear in one picture showing successive stages of the action. 

Although the microflash units have a relatively short range and 
angle of spread, helicopter rotors up to 38 ft across have been success- 
fully photographed during rupture. To accomplish this end, the 
blades were painted white, a fluoride-coated //2. 5 lens was used on the 
camera, and the fast fluorographic film developed to completion. 
Ten minutes in straight D-72 with an antifoggant added is not un- 
usual to bring up an image adequate for printing. 

In the case of the helicopter propeller rupture, a wire was cemented 
to the blades and brought out through a slip-ring device normally 
used to connect strain gauges to recording instruments. The wires 
were connected to a transformer and battery in series. The second- 
ary of the transformer was close to and connected into the input 
of the triggering amplifier. The transformer enabled the circuit 
comprising the rotor and slip-ring to retain a low impedance arid so 



228 P. M. THOMAS AND C. H. COLES Voi 46, No. 3. 

be relatively free from pickup disturbances. The rupturing of the 
rotor broke the wires and initiated the sequence of flashes. 

Because the camera with its fast lens and film was set for time 
exposure to catch the moment of rupture, the whole propeller test 
laboratory had to be darkened completely. Windows that could 
not be covered in the enclosing structure made it necessary to per- 
form the test only after darkness fell. One of the first pictures made 
with this equipment shows pieces of the fabric blade flying away 
from the rotor. The success of the results obtained so far has war- 
ranted the building of a new sequence flasher of far greater light out- 
put to be permanently installed in the propeller test laboratory for 
continuing research. 

Flash Techniques. In some cases where the action is continuous, 
pictures are required at intervals that do not approach motion picture 
frequency and yet each picture must be made with extremely short 
exposure. An actual case was a helicopter hovering above the 
ground. Pictures of the blades were required to determine coning 
angle and bending. 

An aerial night photographic flash unit was adapted to ground 
operation for this purpose. This unit emitted an extremely powerful 
flash of light whose duration was only 1 /5ooo sec. The flash could be 
repeated 3 times a sec. An aerial camera taking a 5 X 7-in. picture 
had its shutter removed and "its mechanism altered to move the film 
continuously. The lens was set into a focusing mount and the 
camera set upon a Mitchell tripod. A contactor was installed in the 
camera to flash the light every time a fresh 5-in. length of film came 
into position. 

When darkness fell on the flying field and everything was in readi- 
ness, the helicopter pilot was given the signal to make the aircraft 
rise to a hovering position. At the same moment, the camera was 
started and the light was fired by the film metering rollers inside. A 
series of pictures was thus obtained showing the blades of the heli- 
copter sharply defined against a black sky. Measurements could then 
be easily made of the angle and deformation of the blades. 

Individual flash pictures may, of course, be made with the same 
equipment. An example of such an application was the request made 
by the Propeller Laboratory to photograph the successive stages of 
the building up of ice on a propeller. The request stated that this was 
not to be done in the wind tunnel but must be accomplished under 
actual icing conditions in the air. To complicate matters further, 



March, 1^46 HlGH-SPEED PHOTOGRAPHY 229 

it was considered too dangerous to make the flight at night, sufficient 
hazard being encountered during daylight operations in icing clouds. 

The problem was finally solved by the combination of several 
techniques and the development of a new discharge lamp. The 
regular lamp of the night photographic 'unit was replaced with a short 
duration tube. This tube flashed in about 1 /2o,ooo sec. To make the 
picture, a wide-angle camera was constructed which could be operated 
entirely from the rear and so rigid that it would keep its focus despite 
the vibration of the airplane. A contactor was fitted to the shutter 
to fire the flash when the blades were wide open. In this way, the 
effect of daylight would be kept to a minimum with a high shutter 
speed. 

The camera and lamp were installed in the cockpit of the B-25 
Mitchell bomber directly behind the pilot, viewing the blades of the 
propeller perpendicular to their axis of rotation. The success of the 
whole project depended upon 3 factors: (1) overpowering the day- 
light with the flash, (2) a fast enough flash to stop the propeller, and 
(3) sufficient contrast to be obtained between the cloud background 
and the propeller blade to show the latter to its best advantage. 

To achieve the last requirement, color contrast was tried and found 
to be of considerable help. Ansco Color Film was loaded into the 
camera and the propeller blades of the airplane painted bright red. 
The riame of the B-25, Flaming Mamie, was no meaningless term as 
one glance at the flaming color of the propeller would prove. 

The photographs obtained with this equipment showed the white 
ice crystals on the red propeller blade against the blue cloud back- 
ground, a colorful and successful solution to a difficult problem. 

Motion Picture Theodolites. Recording theodolites are used 
effectively at Wright Field for the location in space of moving air- 
craft, parachutes, and slow-moving missiles. The recording theo- 
dolites are essentially motion picture cameras whose azimuth (pano- 
rama from the north point) and site (tilt from the horizontal) are re- 
corded on the film simultaneously with the picture. A clock is also 
recorded for the purpose of matching pictures taken at the same time 
from 2 stations. The theodolites are always used in pairs so that 
triangulation from their 2 positions defines the position of the subject. 
The clocks on the 2 instruments may be synchronized by radio so that 
accuracy in timing is assured. From the observations recorded by 
these theodolites, three-dimensional space graphs may be plotted to 
depict the exact motion of an object in the air. 



230 P. M. THOMAS AND C. H. COLES 

With these instruments, the flight path of a helicopter was re- 
corded and plotted to prove that such an aircraft requires some wind 
to produce vertical ascent. The oscillations and drift of a para- 
chute were also measured. 

Conclusion. Motion picture technique has been put to work in 
the ways enumerated to aid in the solution of engineering problems 
at Wright Field, the experimental center of aircraft development for 
the Army Air Forces. Under the pressure of war, the satisfactory 
solutions to these design problems had to be found quickly. The suc- 
cess which attended the application of these new photographic ana- 
lytical methods was so complete that this work is expanding to an 
ever-increasing extent. 



COLORED TRACE OSCILLOGRAMS* 

L. S. TRIMBLE AND F. W. BOWDEN** 

Summary. Aerial Kodacolor has been found suitable as a cure for trace en- 
tangled, illegible, oscillo graphic recordings. A 5 l /?-in. "width film is available that in 
combination with filtered light beams is productive of distinctly differing color traces 
at recording speeds up to 20 in. per sec. 

It is often found necessary to study a function that changes rapidly 
with time ; a convenient method of measuring this variation is to con- 
vert it to an electrical impulse and record the corresponding variation 
by means of an oscillograph. A simple form of this instrument em- ' 
ploys a sensitive galvanometer capable of causing deflections in a 
beam of light by virtue of being coupled electrically to the variable 
whose time function it is desired to study. In the usual case the light 
beam is reflected from a mirror on a galvanometer and impinges and 
records on a photosensitive film or paper driven uniformly with time. 
The narrow light beam is then deflected normal to this direction of 
travel so that the processed photosensitive material carries a con- 
tinuous line trace representing the function variation with time. 

In many cases several light beams are caused to record on the same 
photosensitive material, and often these beams will be simultaneously 
deflected so that trace entanglement results, making the final study 
of an individual line practically impossible. This condition is ac- 
centuated in aircraft flight test recordings where, because of space and 
weight considerations, as many as 12 oscillating traces are recorded 
on a 6-in. width of photographic paper. Fig. 1 illustrates such an 
entanglement as the result of flight stresses recorded on the Lockheed, 
60-passenger, high-speed Constellation. As can be seen, considerable 
time is required to follow each trace and determine its frequency, 
amplitude, and slope characteristics. 

Mechanical alterations of line width or continuity have not been 
successful in positive identification. Compact light-weight instru- 

* Presented May 17, 1945, at the Technical Conference in Hollywood. 
* * Lockheed Aircraft Corporation, Burbank, Calif. 

231 



232 L. S. TRIMBLE AND F. W. BOWDEN Vol 46, No. 3 

ments capable of equivalent individual recording are not currently 
available. One method of simplifying this problem is to utilize a 
color sensitive and reproductive photo-recording material thus allow- 
ing the reproduction of traces as separate and distinct hues, preferably 
upon a white background. The oscillograph needs altering only in 
the proper filtering of the light beams and the substitution of a suit- 
able color sensitive photographic material for the orthochromatic type 
of bromide paper normally used. 



r. /v 

V W \ ff 



i^M^,*'- ' 




'; ^y^'f^'f- 



*^v^ ^ ^v^^!v*v^^^^^* l i*H'**<V- 

#^4v :: "''""~'~'' : * Vr ' 



FIG. 1. Black and white entangled traces. 

A gelatin or glass filter having transmission characteristics similar 
to one of those shown in Fig. 2 is placed in each beam so that the 
light passes through it only once.- The exact location of the filter will 
depend upon the instrument used as well as the optical system of the 
instrument. Since there are but 6 major colors in the visible spec- 
trum, and there are 12 traces to filter, sharp, clear, visual color dis- 
tinction between all lines is difficult to attain. Two blue lines, for 
instance, differing in density or differing but slightly in hue, will 
form closely identical traces as one beam is rapidly modulated, thus 
altering photographic exposure and thereby color density, and to some 



March, 1946 



COLORED TRACE OSCILLOGRAMS 



233 



extent, hue. In addition, certain limitations in the color reproduc- 
tion mediums available at the present time serve to narrow the re- 
producible color range. It is possible, however, to select Wratten 
niters differing in spectral transmissions so that corresponding vari- 
ations in hue will be evident on the recording medium. Widely 
spaced traces may be in duplicate hues since the probability of <their 
complete intermodulation would be slight. The quality and thick- 
ness of the niters should be chosen so that the over-all transmissions 
are of about the same order, thus insuring light intensities within the 
latitude rangte of the recording medium. 




500 550 

WAVELENGTH (m/u) 



FIG. 2. Filter characteristics. 

Flight test records, particularly on new model airplanes, are often 
secured in a few moments of flight time, but represent many hours of 
planning. Modifications or additional tests are considered only after 
interpreting first performance data, so that speed of processing records 
as well as reading the records is important. For this reason, those 
color photographic products easily processed by the customer have 
received the most attention. Most of the available photographic 
materials comprise 3 light sensitive emulsions which record the 3 
color aspects of the impinging light beam. During processing of 
certain of these materials the exposed silver is reduced in the normal 



234 



L. S.* TRIMBLE AND F. W. BOWDEN Vol 46, No. 3 



manner followed by the coupling of the partially oxidized form of the 
developing agent with substances incorporated in each emulsion such 
that dye images are formed of a hue complementary to the wave- 
length band responsible for the emulsion exposure. A blue-violet 
beam, therefore, will produce a yellow line on the photographic 
medium, whereas a green beam will produce a magenta line. Cer- 
tain other color products are designed to be treated by a reversal 
process, such that the reproduced hue will substantially match that 
of the filtered light beam. 



2 
2.4 
2.2 
20 

f;: 

~" 1.4 
| 1.2 
S 1.0 

I- 

0.6 
0.4 
0.2 
0.0 




-2.5 -2.2 -1.9 



-i.o -0.7 -0.4 -o.i 0.2 

LOG. EXPOSURE {MCSi 



FIG. 3. Comparison of available recording materials. 

Comparison of color paper and film speeds was made by means 
of sensitometric lib exposures at 3000 K, a color temperature speci- 
fied for product color balance; processing was conducted utilizing 
packaged chemicals supplied with the recording material in accord- 
ance with the manufacturer's instructions. Fig. 3 shows typical 
sensitometric curves of the paper and film recording mediums avail- 
able. Diffuse density, transmission or reflection, is plotted against 
log exposure in the normal manner. The slope of the straight line 
portion of the curve provides a comparison of contrast; and for a 
given density and contrast, a change of 0.3 log exposure represents 
100 per cent, or one stop, change in exposure. 



March, 1946 COLORED TRACE OSCILLOGRAMS 235 

Data on 2 commercial blue-sensitive papers are shown for reference, 
a chloride type of contact paper, and a bromide projection or en- 
larging paper. It can be seen that the commercial black-and-white 
oscillogram paper is about 2 stops faster than the projection paper, 
whereas the available color paper is about 4 stops slower than the 
oscillogram paper. By limiting the bromide concentration in the color 
developer of the nonreversal process, it is possible to gain 2 to 3 stops 
in speed, as shown by the dotted curve, without appreciable gain in 
fog level. This allows the complementary color rendition of line 
images at a linear paper speed of one to 2 in. per sec. Although this 




FIG. 4. Developing pans and chemicals. 

is satisfactory, the margin of speed is not great, the color distinction 
is not particularly good, and the thick paper tends to bind and surface 
crack in passing through the oscillograph. 

Photographic films are faster than any of the photographic papers 
available; they must be viewed by projection or over a light box, but 
have substantially increased exposure latitude, are capable of good 
color separation, and have the ability to be duplicated in the form of 
paper prints for detailed study or report purposes. 

Satisfactory results were obtained up to a linear recording speed of 
20 in. per sec through the use of Aerial Kodacolor supplied in SVVin. 
by 40-ft rolls. The 6-w tungsten lamp in the oscillograph, a Miller 
model H, provided a satisfactory light source, and the individual light 



236 L. S. TRIMBLE AND F. W. BOWDEN 

beams were filtered using Wratten gelatin squares Nos. 23, 29, 
K2, 34, 38, 40, 45, 47, and 61N. The rolls were processed in about 
l x /2 hr by winding the film on a Steinman-type spiral and immersing 
in 4 x /2 gal of the processing solutions prepared from the chemicals 
supplied with each roll, Fig. 4. Exposure for reversal was accom- 
plished with the film on the spiral by traversing the spiral path be- 
tween the film planes with a glowing relatively heat-free light source 
comprising a mercury arc in the form of a 3 /ie-in. glass 7-tube. Air 
circulation from a fan hastened drying of the film on the spiral. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 



American Cinematographer 

27, 1 (Jan., 1946) 

Orthicon Pickup Tube for Television Cameras (p. 6) 
Pointers on Use of New Ansco 16-Mm Color Film (p. 7) 
Automatic Follow-Focus Device for Use in Cinema- 
tography (p. 8) 

Sixteen Goes Hollywood (p. 12) 
No Cherry Blossoms in a Factory (p. 16) 
Using Your Movie Camera as a Motion Picture Step Prin- 
ter (p. 24) 

British Kinematograph Society, Proc. Sub-Standard Division 

(1944-45) 

Sub-Standard Motion Picture Practice (p. 2) 
Professional Sub-Standard Projection (p. 11) 
Screen Brightness in Sub-Standard Projection (p. 16) 
American Standards for 16-Mm Service Projectors (p. 19) 
Sound Recording for Sub-Standard Films (p. 24) 

Electronic Engineering 

18, 215 (Jan., 1946) 

A Continuous Film -Recording Camera for Use with Stand- 
ard Cathode-Ray Oscilloscopes (p. 10) 

Ideal Kinema, The 

12, 126 (Jan., 1946)' 
Westrex Sound System (p. xxv) 

International Photographer 

17, 12 (Jan., 1946) 
The Strobolight (p. 9) 
Rotocolor is Something Different (p. 11) 



J. T. STROHM AND 
W. G. HECKLER 
R. FERNSTROM 
W. WISE 

J. R. OSWALD 



I.D. WRATTEN 
W. HINE 
G. H. SEWELL 
R. H. CRICKS 
N. LEEVERS 



A. H. SIMONS 



G. S. APPELGATE 



F. J. SCHERSCHEL 

R. FERNSTROM 

237 



238 CURRENT LITERATURE 

International Projectionist 

21, 1 (Jan., 1946) 
A Post- War 16-Mm Projector: The Ampro Premier 10 

(p. 7) L. CHADBOURNE 

Sharp Heat Reduction, Better Color Rendition Claimed for 

New Glass (p. 12) 
Projectionists' Course on Basic Radio and Television Pt. 

19, Vacuum Tubes (p. 20) M. BERINSKY 

The Stratovision System for Television, FM Pt. 2 (p. 24) C. E. NOBLES 

Technique Cinematographique, La 
10, 12 (Dec., 1945) 

Standards of the Characteristic Dimensions of Auditoriums 
(Russian Standards) (p. 221) 

Television 

3, 1 (Jan., 1946) 
Film Projection Equipment (p. 13) J. L. CADDIGAN 




59th SEMI-ANNUAL TECHNICAL CONFERENCE 

Hotel Pennsylvania, New York 

May 6-10, 1946 

Directory of Committee Chairmen 

Atlantic Coast Section and Local Ar- 
rangements FRANK E. CAHILL, JR., Chairman 

Registration and Information W. C. KUNZMANN * 

Luncheon and Dinner-Dance Commit- 
tee E. I. SPONABLE, Chairman 

Hotel and Transportation O. F. NEU 

Membership and Subscription Commit- 
tee JAMES FRANK, JR., Chairman 

Ladies Reception Committee Hostess MRS. O. F. NEU 

Papers Committee. . , y BARTON KREUZER, Chairman 

C. R. DAILY, Vice-Chairman 

Publicity Committee HAROLD DESFOR, Chairman, assisted by 

LEONARD BIDWELL 

Projection Programs 35-mm H. F. HEIDEGGER, Chairman, assisted 

by Members New York Projection- 
ists Local 306 
16-mm J. E. STEOGER 

HOTEL RESERVATIONS AND RATES 

No hotel room reservation cards will be mailed to the membership for this Con- 
ference. Therefore, members and others must book and reserve desired room ac- 
commodations early and direct with Joseph Troise, Front Office Manager, Hotel 
Pennsylvania, New York 1, N. Y., prior to April 20. Mention that the reserva- 
tion is in connection with the SMPE Technical Conference. No rooms will be 
assured or available unless confirmed by the hotel management. 

239 



240 FIFTY-NINTH TECHNICAL CONFERENCE Vol 46, No. 3 

Note: Out-of-town members who can schedule their New York arrival for 
Sunday, May 5, are more apt to get immediate room assignment on this date 
than if arrival is on Monday, May 6. 

The following per diem room rates, European plan, are extended to SMPE 
members and guests when booking accommodations direct with the Hotel Penn- 
sylvania : 

Room with bath, one person $3 . 85, $4 . 40, $4 . 95, $5 . 50, $6 . 05, $6 . 60 

Room with bath, 2 persons, double bed $5 . 50, $6 . 05, $6 . 60, $7 . 15, $7 . 70 

Room with bath, 2 persons, twin beds $6,60, $7. 15, $7.70, $8.25, $8.80 

Parlor suits for one or 2 persons $10. 00, $11 . 00, $13 . 00, and $18. 00 

REGISTRATION 

The Conference registration headquarters will be located on the 18th floor of 
the hotel adjacent to the Salle Moderne, where all business and technical sessions 
will be held during the 5-day Conference. Members and guests are expected to 
register. The fee is used to defray Conference expenses. 

TECHNICAL SESSIONS 

SMPE members and others contemplating presentation of papers at this 
Technical Conference can greatly assist the Papers Committee in the early as- 
sembly of the program by mailing in title and author of papers together with an 
abstract by April 1, 1946. Complete manuscripts should be sent to the Chair- 
man or Vice-Chairman of the Papers Committee not later than April 15. 

Only through your earnest co-operation will it be possible to draft and announce 
the papers program prior to the opening of the Conference. 

SMPE GET-TOGETHER LUNCHEON 

The Society will again hold its regular pre-war social functions, and accordingly 
a Get-Together Luncheon is scheduled in the Penn Top (formerly the Roof Gar- 
den) on the 18th floor of the hotel, on Monday, May 6, at 12: 30 P.M. Ladies are 
invited to attend this luncheon. Tickets must be procured at the registration 
desk prior to noon on May 6, so that adequate hotel accommodations may be pro- 
vided accordingly. 

The Board of Governors cordially invites the holders of Dinner-Dance tickets 
to spend a social hour with the Board in the hotel Georgian Room between 7 : 15 
P.M. and 8:15 P.M., on May 8, preceding the Conference dinner. (Refreshments.) 

The informal Dinner-Dance (dress optional) will be held in the Georgian Room 
promptly at 8:30 P.M., on May 8. Dancing until 1:30 A.M. 

Note: It is imperative that Dinner-Dance tickets be procured and table reser- 
vations made at the registration headquarters prior to noon on May 8. Your 
earnest co-operation with the Arrangements Committee is requested. 

LADIES' PROGRAM 

A reception parlor will be provided in the hotel for the ladies' daily get-together 
and open house. The ladies' entertainment program will be announced later. 



March, 1946 FlFTY-NlNTH TECHNICAL CONFERENCE 
MOTION PICTURES AND RECREATION 



241 



Conference identification cards issued to registered members and guests will 
be honored at New York deluxe motion picture theaters which will be listed in later 
issues of the JOURNAL. 

Those interested in other entertainment while in New York should consult the 
hotel information bureau, or the SMPE registration headquarters. 



Monday, May 6, 1946 

Open Morning. 
9: 30 a.m. Hotel, 18th Floor: Registration. Advance sale of Luncheon and 

Dinner-Dance tickets. 

12 :30 p.m. Hotel Penn Top: (formerly Roof Garden) , 18th Floor: Get-Together 
Luncheon. (Eminent Speakers.) Note: Luncheon tickets must 
be procured before noon on May 6, at the registration desk. 
2 : 00 p.m. Salle Moderne: Opening session of the Conference. Business and 

Technical Session. 
8:00 p.m. Salle Moderne: Evening Session. 

Tuesday, May 7, 1946 

9: 00 a.m. Hotel, 18th Floor: Registration. Advance sale of Dinner-Dance 

tickets. 

9: 30 a.m. Salle Moderne: Morning Session. 
2:00 p.m. Salle Moderne: Afternoon Session. 
Open Evening. , 

Wednesday, May 8, 1946 

Open Morning. 
10:00 a.m. Hotel, 18th Floor: Registration. Advance sale of Dinner-Dance 

tickets. 

2:00 p.m. Salle Moderne: Afternoon Session. 
7:15 p.m. Georgian Room (Reception Foyer): A social hour with your Board 

of Governors preceding the Dinner-Dance. (Refreshments.) 
8:30 p.m. Georgian Room: Fifty-Ninth Semi- Annual Technical Conference 
Dinner-Dance. Social get-together, entertainment, and dancing 
until 1:30 A.M. 

Note: Tickets must be procured and tables reserved prior to noon on 
May 8, for this function. 

Thursday, May 9, 1946 
Open Morning. 



2:00 p.m. 
8: 00 p.m. 



Salle Moderne: 
Salle Moderne: 



Afternoon Session. 
Evening Session. 



242 FIFTY-NINTH TECHNICAL CONFERENCE 

Friday, May 10, 1946 

9: 30 a.m. Salle Moderne: Morning Session. 

2 : 00 p.m. Salle Moderne: Afternoon Session. Adjournment of the Fifty- 
Ninth Semi-Annual Technical Conference. 

Note: All sessions during the 5-day Conference will open with an interesting 
35-mm motion picture short. 

IMPORTANT 

Those desiring hotel rooms must book their accommodations direct with the 
Hotel Pennsylvania management prior to April 20, which are subject to cancella- 
tion prior to May 1." 

Owing to the acute travel conditions, it is imperative that out-of-town members 
and guests who contemplate attending the May Technical Conference consult 
their local railroad passenger agent regarding rail and Pullman accommodations, 
within the existing Pullman reservation period. 

W. C. KUNZMANN 
Convention Vice-f 'resident 



EMPLOYMENT SERVICE 
POSITIONS OPEN 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation. 18 
West 48th St., New York 19, N.Y. 



Position available for Optical Designer, capable of handling the calcula- 
tion and correction of aberrations in photographic and projection lens 
systems. Junior designers or engineers will be considered. Write 
fully giving education, experience, and other qualifications to Director 
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi- 
cago 45, 111. 



POSITIONS WANTED 

Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write J. I. K., 
354 Ninth Ave., New York 1, N.Y. 



Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 APRIL, 1946 No. 4 

CONTENTS 

PAGE 
The Illusion of Depth in Motion Pictures 

H. T. SOUTHER 245 

The Theory and Practice of Lighting for the Camera 

H. T. SOUTHER 254 

Westrex Standard Sound Film Reproducer 

G. S. APPELGATE AND J. C. DAVIDSON 272 

Westrex Master Sound Film Reproducer 

G. S. APPELGATE AND J. C. DAVIDSON 278 

American Standards on Motion Pictures 284 

Report of the Membership and Subscription Committee 310 

Officers and Governors of the Society 312 

Committees of the Society 315 

Constitution and By-Laws of the Society 322 

Journal Award and Progress Medal Award 333 

Report of the Treasurer 336 

Society Announcements 337 



Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCItTY of MOTION PICTURE ENGINEERS 

MOTCL PENNSYLVANIA NCW YOKK I. N-Y' Tt. PCNN. 6 O62O 

HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 

* President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 

* Past- President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
*Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
** Engineering V ice-President: JOHN A. MATJRER, 

37-01 31st St., Long Island City 1, N. Y. 
^Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
**Financial Vice-P resident: M. R. BOYER, 

350 Fifth Ave., New York 1. 
* ''Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
^Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
* Treasurer: EARL I. SPONABLE, 

460 West 54th St., New York 19. 

Governors 

*fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
**WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
"tHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairman, Atlantic Coast Section. 
**Term expires December 31, 1947. ^Chairman, Pacific Coast Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol. 46 APRIL, 1946 No. 4 

THE ILLUSION OF DEPTH IN MOTION PICTURES* 
HOWARD T. SOUTHER** 



Summary. Motion pictures are not realizable as such until they are perceived 
by the brain through the eye. The manner in which the eye operates is dealt with, 
particularly those functions having to do with perceiving objects as solids. The in- 
telligent employment of motion picture taking equipment now in use, in making it 
conform to the requirements of the eye, can result in an important contribution to three- 
dimensional perception from the screen. Some 15 aids in promoting the illusion of 
depth are dealt with in this paper. 

Foreword. Leonardo da Vinci said that pictures were the result 
of "giving corporeal shape to the three dimensions on a flat surface." 
In the execution of his masterpieces, Leonardo the artist used Leo- 
nardo the scientist to fix his design, to project his true perspectives, 
to mix his colors, and to imprison light like Ariel in his web. This 
paper presumes to deal with the methods by which the three-dimen- 
sional illusion is evoked by calling attention to some 15 contributing 
factors. 

But before going further, it is well that we understand that any 
pictorial effect is not realizable as such until perceived by the brain 
through the eye. In a simple and general way, the camera, projector, 
and associated equipment serve as a delay mechanism for the pres- 
entation to an audience of an occurrence which it is desired that they 
should see. We must keep in mind at all times that the eye and its 
peculiarities of perception should govern the steps by which this delay 
is accomplished. 

Two broad corrections in lighting treatment are needed to achieve 
this delay. The first is a considerable compression of the range of 
light intensities falling on the subject. This is necessary in order 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Lieutenant, Signal Corps, Signal Corps Photographic Center, Long Island 
City, N. Y. (Twentieth Century-Fox Film Corporation, Hollywood.) 

245 



246 



H. T. SOUTHER 



Vol 46, No. 4 



that the exposure limits of the camera may be met. These are much 
less than those of the eye and generally call for an increase in the 
normal amount of light present. Second, a correction for color is 
necessary. The eye and the film are not linear with respect to each 
other. For instance, yellow affects the eye more violently than the 
film; blue affects the film more violently than the eye. There are 
many other differences which call for compensation. We shall take 
up these points of variance more completely later. 

The Illusion of Three Dimensions. One of the aims of motion 
picture presentation is to achieve the illusion of reality. Optically 





A B 

FIG. 1. A The shape of contours; B Overlap. 



speaking, we must duplicate wherever possible in the minutest 
detail the actual experience of vision. Paradoxically, the require- 
ments of perfection should not be too strongly emphasized. The 
human being is capable of considerable psychological adjustment to 
his environment in general; in this case, to the screen in particular. 

The screen portrays its subjects many times oversize. We all 
realize that there is an easy adjustment to this seeming gigantism. 
There is also the adjustment which permits comfortable observation 
from the unnatural angle of the viewer. 

In viewing a motion picture, the eye observes the scene in only 2 
planes actually. By employing the proper technique, an illusion of 
three dimensions may be evoked. 

Methods of Portraying Solidity. The illusion of three dimensions 



April, 1946 ILLUSION IN MOTION PICTURES 247 

is built up upon the screen in a number of different ways. The 
following inethods are employed universally at present : 

The Shape of Contours. We view in Fig. I A the outline of a vase. 
The drawing means to suggest merely the feeling of an enclosure 
separated from its background by only a line. 

Overlap. In Fig. IB we observe a slight increase in the feeling of 
form. The outlines of the 2 vases complement each other by virtue 
of the overlapped position. 

Cast Shadows. The cast shadows of the 2 vases, in Fig. 2, on the 




FIG. 2. Cast shadows. 

background and the shadow of the first vase on the second have con- 
tributed tremendously to the feeling of form. Note also that the 
shape of the shadows has a contributing effect. Interest increases. 
Observe the contrast of light and shade. As the shadows are 
lightened the beneficial effect decreases. Heavy contrast results in 
more powerful delineation and perception of rounded form. 

Perspective. The simulation of an actual viewpoint and the normal 
decrease in size because of distance have resulted in an increased 
effect, seen in Fig. 3. We are impressed psychologically by the 
duplication of an effect observed in everyday life. 



248 H. T. SOUTHER Vol 46, No. 4 

Reflections. The shape and position of the reflection of the vase in 
the water below in Fig. 4 is something we would expect in normal 
existence. This is another duplication of the actual which increases 
interest and illusion. 

Elevation and Light Reflections. In Fig. 5 the form of the vase has 
changed in shape because of the changed viewpoint. The phenome- 
non of foreshortening is experienced in graphic form. 

The tempering of the shadow by reflected light produces another 
simulation of reality. Daily experience again repeats in this figure. 




FIG. 3. Perspective. 

All of these phenomena help in the three-dimensional illusion to a 
marked degree. The most important aid to this illusion, none the 
less, would be one which we miss in motion pictures almost without 
exception; that is, the function of the eyes in a stereoscopic manner. 
A more complete understanding of this important function of the eye 
may be achieved if we engage in a digression on the phenomenon of 
vision. 

Definition of Sight. Sight may be termed the perception by 
the brain through the eye of varying intensities of wavelengths of 
light, radiated or reflected from a substance or object. Radiation 
may take place from generation within the object, in which 
case it incandesces, or glows, or by reflection of light by an objec- 
from the original source. Varying intensities of reflection from port 



April, 1946 



ILLUSION IN MOTION PICTURES 



249 



tions of the object impinge themselves upon the retina of the eye, and 
cause the object to assume form in our brain. 

Sensitivity of the Eye. The sensitivity of the eye to different 
amounts of light can be measured. In the same manner, the sensi- 
tivity to various degrees of light on motion picture films can also be 
measured. However, the human eye is a wonderful device. It is 
much more sensitive in ordinary ways than the film. 

The eye can detect changes of light of one in a million. The camera 
cannot. The ability of the eye to define an object is infinitely more 
acute. When we see a motion picture we must look through the eye 




FIG. 4. Reflections. 



of the camera. Whereas the retina of the eye consists of microscopic 
cells over 8,000,000 in number to the square inch, the finite granular 
structure of the film and the focusing limits of the camera decrease the 
definition of the object whose reflected light has caused an exposure 
upon its surface. This phenomenon results in a distortion of the per- 
ceived image. The sensitivity of the film to color does not at present 
correspond with that of the eye. This results in a further distortion. 
Distortion by the Eye. The knowledge of the manner in which 
these distortions take place, and in what degree, are important in 
our work. Distortions necessarily need not be bad. If they are 
controlled they can be very useful and may contribute materially to 
the artistic effect of a scene. A paper flower may be distorted so as to 



250 H. T. SOUTHER Vol 46, No. 4 

/ 

appear more beautiful than a real one by proper lights and a camera. 
The converse is also true. A real flower may be distorted by bad 
lighting and camera operation so as to appear like a paper one. The 
difference in achieving the desired effect and the opposite bad effect 
lies in the proper use by the various technical departments of their 
equipment. 

The Analogue of the Camera and the Eye. The camera and the 
eye partake of a number of mutual properties. The first of these 
is the physical one of construction. Fig. 6 shows that both the 
camera and the eye possess lenses which focus or gather the light rays 




FIG. 5. Elevation and light reflections. 

and concentrate them upon the desired spot. In the eye this spot is 
the fovea of the retina. In the camera it is the film. If the eye and 
the camera reacted in the same manner in every way, our problems 
would be simplified. But it is the difference between the eye and the 
camera which concerns us. The operations and adjustments neces- 
sary to make the camera react like the eye, or in any manner which we 
may desire, are the reasons for the further study on this subject. 

Convergence and Divergence of the Eyes. The higher verte- 
brates are able to converge the axes of the eyes on nearer points. 
This enables the images of the point to coincide with the central 
spots of both retinas. This ability of the eye, and others, is associ- 
ated with the development of the higher faculties of the mind. The 



April, 1946 



ILLUSION IN MOTION PICTURES 



251 



normal position of the human eyes is convergent or parallel, but it is 
possible also for the eyes to diverge. 

The movement of the eyes is complex. When they move together 
to one side or the other, up or down in a vertical plane, there is no 
rotation of the optical axes, i. e., no torsion. When the visual plane 
is elevated and the eyes move to the right, they rotate to the right. 
When they move to the left, they rotate to the left. When the visual 




FIG. 6. The analogue of the camera and the eye. 



plane is depressed and the eyes turn to the right, they rotate to the 
left, and vice versa. We constantly evaluate these complex muscular 
stresses, and through experience we interpret size, shape, and distance 
of objects. We must believe from this that a baby newly born must 
learn to see in 3 planes. They learn how far to reach for an object 
through evaluating muscular stress in the eye, and through actually 
evaluating the physical effort required in comparison to reach for it 
with their arm. 

Accommodation by the Eye. Objects at different distances can- 
not be seen clearly at the same time. However, by interpreting the 



252 H. T. SOUTHER Vol 46, No. 4 

eye movements as the point of sight is focused forward and back- 
ward, the intellect automatically appraises size, form, and the dis- 
tance of each object. This is the result of the ability of the eyes to 
focus upon a particular object and the ability of the observer's mind 
to determine from focus, and from previous experience in viewing 
other objects, the approximate distance of the object viewed from the 
eyes. This is part of the factor of accommodation. 

The combination of convergence and accommodation, carried out 
unconsciously and automatically, produces the major depth effect. 

Norling 1 says that the fundamental problem in projecting three- 
dimensional pictures is that of "projecting a 'right eye' image which 
will reach the right eye, and projecting a 'left eye' image which will 
reach the left eye." 

If we were to assume that each eye was capable of independent 
scanning in the human being, the necessity of a particular picture 
for the right eye and a particular picture for the left eye would be 
eliminated. We know that stereoscopic effects are achieved by the 
camera today occasionally in the monoptical manner. 

The premise is proved further by medical research. Ives 2 says 
that the old-time stereopticon photos resulted in a strain on the part 
of the eyes when viewed for more than a short period of time. This 
would tend to prove the idea of independent, or monoptical scanning. 
This would tend to prove, also, that the 2 eyes, scanning point for 
point at the same time, are doing an unnatural thing when they 
observe views in a stereopticon which keep them from scanning 
independently. 

Scanning. The eyes constantly scan a scene being viewed. 
When we dolly a camera, or when we follow with our eyes the actor 
as he walks across the screen, we are increasing the feeling of three 
dimensions very materially by scanning. We are injecting arti- 
ficially one of the stereoscopic effects of which the eyes are capable 
when viewing the object in real life. 

Monocular Stereoscopic Vision Versus Binocular Stereoscopic 
Vision. Now, a great deal is not known of the exact manner in 
which the eye functions. Most authorities agree that stereoscopic 
vision implies binocular vision. But this point is open to question. 
If this were true, it would hold that animals, the axes of whose eyes 
are spaced 180 degrees apart, would have considerable difficulty in 
perceiving near objects because of the inability to converge. The 
actions and acuity of a deer, for instance, as far as distance is 



April, 1946 ILLUSION IN MOTION PICTURES 253 

concerned, would preclude such restriction. In this case, the 
animal would have no more advantage than the human in viewing 
the present motion picture with only those previously exposed 
aids to viewing depth. This gives rise to the idea that distance can 
be realized very well monoptically . 

The writer is inclined to believe that those stereoscopic effects 
achieved at times accidentally on the screen have as their basis 
a scanning operation by the camera. In effect, the camera 
duplicates a normal function of the eyes. 

This, if true, would tend to show that stereoscopic vision is not 
entirely, if at all, a matter of effect achieved binocularly. Among 
other things, it means that we have room for unlimited improvement 
in our work with present equipment. Perhaps our problems are 
largely concerned with the refinement of technique. 

Psychology of the Eye. Suggestion plays a great role in the art 
of seeing. Always we must hold in mind that we see with our brain 
through the eye. What the mind believes, the eye will see. We 
may hide the method, and the eye will believe the result, however 
achieved. 

It has been the writer's intention in the foregoing paragraphs to 
show that a knowledge of the biology and psychology of seeing are 
very important. The intelligent use of equipment in conforming to 
the requirements of the eye will result in a great step toward our 
goal "corporeal shape to the three dimensions on aflat surface" 

[Ed. Note. A 16-mm Kodachrome motion picture was shown illustrating a 
number of the various points discussed in the foregoing, emphasis being placed on 
the relative apparent motion of objects when the camera was moving. The 
demonstration is referred to in the Discussion section following.] 

REFERENCES 

1 NORLING, J. A.: "Three-Dimensional Motion Pictures," /. Soc. Mot. Pict. 
Eng., XXXIH, 6 (Dec., 1939), p. 612. 

2 IVES, CHARLES E., Eastman Kodak Company. 



THE THEORY AND PRACTICE OF LIGHTING FOR 
THE CAMERA* 



HOWARD T. SOUTHER** 

Summary. It is the purpose of proper lighting to mold the dimensions, transmit 
the plane, and give to motion pictures the necessary mood and beauty. The screen 
image is a modified light reflection of reality, and light is the main source of photo- 
graphic construction. This paper deals with the all important necessity of successful 
production, the exposition of lighting as a method. The theory and its application in 
practice are discussed. The employment of the principles of illusion in achieving 
photographic presentation is considered. 

Foreword. The apparent mission of the cameraman is to flatter 
the senses with an illusion of reality. His secret and true aim is to 
purvey a psychical effect. The various methods of distortion, 
whether they be a soft presentation of the image or the sharp trans- 
mission of linear optics, all require subjective and stylistic motiva- 
tion. If initiative and imagination are not used boldly, the camera 
loses its significance and means of expression. It becomes merely a 
tool of technique and artisanship. 

But here we are faced with a problem. Although we may be 
artists at heart, we must be technicians in fact, in order that we may 
employ the technology of our art to paint the soul of the subject. 
And so the paragraphs which follow will seem intricately and triv- 
ially concerned with processes. Yet even miracles must have proc- 
esses of some kind, however instantaneous. 

In a given object, we may concede that the linear and graphic de- 
tails are inherent in the thing itself. But it is the purpose of proper 
lighting to mold the dimensions, to transmit the plane, to provide 
spatial depth, and to give to the picture necessary mood and beauty. 
Taking into account the laws of optics and photographic transmission, 
the screen image is a modified light reflection of reality. Light is 
the main source of photographic construction. Without its primary 
organizational activity the screen image is impossible. 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Lieutenant, U. S. Signal Corps, Signal Corps Photographic Center, Long 
Island City, N. Y. (Twentieth Century-Fox Film Corporation, Hollywood.) 
254 



THE THEORY OF LIGHTING FOR THE CAMERA 255 

Lighting Theory. Photographically speaking, there are 2 types 
of lighting, each producing a different visual effect : 

(1) Direct, concentrated, or "hard" lighting, which produces a harshly con- 
trasting distribution of light and shade. It reveals sharply the details of the ob- 
ject and the outline of its contours. It transmits the volume of the object through 
linear definition of its perspective. This type of illumination implies lighting 
from a, very intense, single-point source. 

(2) Diffused "soft" lighting, which reveals an easy plastic distribution of light 
and shade. Deep, bright, clear lights are almost entirely absent, while half-tones 
predominate in the image. It suggests space in the form of air as vaguely plastic, 
fluid. It is the tool of aerial perspective, and tends to pictorial beauty with an 
evanescent quality. This lighting method implies illumination from a multiple- 
point light source, each point having a relatively low intensity. 

The General Plan of Lighting for the Camera. The basic idea 
in lighting for the camera is to cause the object being photographed to 
register on the film a different intensity of light from as many planes 
of the object as is possible. This plan may be modified by the com- 
positional and emotional requirements of the scenario. The vi- 
gnette, the silhouette, and the various demands of individual "shot" 
motivation can make necessary wide deviation from this precept. 
However, we shall maintain this basic idea because it develops for us 
the all important necessity of successful motion picture production 
method. 

Development of the Lighting Method. One of the most impor- 
tant means of accomplishing the cinematic illusion is through light- 
ing our object with the thought in mind that it is a series of planes. 
These planes are reflecting surfaces. Apparently each surface forms 
a projecting image when contrasted with a light of different intensity 
than that of the neighboring surface. Through the progression of a 
number of these seemingly opposing planes our conciousness builds 
the illusion on film of the third dimension when reality discloses only 
two. 

Because the head is a commonly accepted subject for photography, 
and in view of the fact that it contains a complex series of planes, pro- 
tuberances, and hollows, we shall use this difficult symbol for illus- 
tration. The mediums with which we illuminate this subject will be 
controllable sources of light. Some of these lighting elements will 
be beams and others will be lamps providing soft broad sources. 
Usage has given these units various names which are indicative of 
their function. 



256 



H. T. SOUTHER 



Vol 46, No. 4 




a 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 257 

The Main-Light or Key-Light. The interception of light by an 
object in everyday life is from above. Our key-light simulates the 
direction of the light source. Usual composition places the object 
in a position facing the light source, and at an angle of 45 degrees 
below and around it. The standard lighting set up is shown in Fig. 1. 
Later we shall take up the use of the other lights shown one by one. 

This results in the casting of certain shadows. It is well that we 
disgress for a moment and consider that there are shadows of 3 
general classifications, seen graphically in Fig. 2 : 

(1) Primary Shadows which are shadows on the object itself, consisting of 
those portions which receive no light from the source. This is the simplest form of 




FIG. 2. .Diagram showing primary, secondary, and tertiary shadows. 

shadow manifestation, but contributes materially to the modeling of the subject. 

(2) Secondary Shadows Reference to the figure reveals this shadow to be the 
one thrown on the background by the interception of the object itself with the 
light source. This shadow adds to the modeling effect, but is not nearly as im- 
portant as those of the primary type in achieving an illusion of depth. 

(5) Tertiary Shadows which are those caused on the object itself by other ob- 
jects intervening between itself and the light source. The importance of this 
shadow varies with the circumstances and form of the intervening objects. In 
those cases where it may be employed, modeling is materially enhanced. It can 
be seen that these shadows can be used to relieve a broad expanse illuminated by 
the key-light and assist in perceiving irregular contour. 

Observe in Fig. 3 the phenomenon caused by the key-light of 
primary and secondary shadows under the eyes, the nose, on the 



258 H. T. SOUTHER Vol 46, No. 4 

neck, and part of the chin. When employed in this manner, the 
key-light assumes the following attributes : 

(1) It is generally the hottest or warmest light. 

(2) The key -light simulates the light source. 




FIG. 3. Shadows which are a resulting phenomenon of the 
key-light. 

(5) It is that light which concentrates the attention of the audience. 

(4) It determines by its intensity the general tonal accent of the composition. 

In exteriors, this light might be the actual sun. In some cases 
this light may impinge from the extreme side angle. Sometimes it 
may come as a three-quarter back-light. This latter condition will 
not so clearly reveal the character of the image in a physical way as 
in an emotional way, after the manner of an effect. 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 



259 



The Back-Light. The back-light is placed in some region more 
or less behind the object. Physical requirements of the element 
placement in the shot most often require that this light be situated 
up and out of the range of the lens. 




FIG. 4. Pseudo back-lighting the use of a spot on the 
background to "blow out" the head contours. 



At other times, it may be advantageous, under certain set condi- 
tions, to hide the back-light directly behind the subject's head. In 
other cases, it may be necessary to resort to a method known as 
pseudo back-lighting; the contrast of a brilliant spot on the back- 
ground with the darker edge of the head area, seen in Fig. 4. 

This last method can be used when the set placement is such that a 
back-light can not be positioned upon the subject. 



260 



H. T. SOUTHER 



Vol 46, No. 4 



The use of back-lighting presents the linear outline of the image. 
The contour is sharply defined. Used alone, the frontal plane as- 
sumes the character of the silhouette, a dark expanse of low tonal 
value. 




FIG. 5. The effect of employing fill -light. 

Front-Light or Filler (Fill-Light). This light is of the uncon- 
centrated type. Fig. 5 will show that its broad, vague-point source 
serves to eliminate small creases and shadows. It has the effect of 
flattening out and making less apparent irregularities. 

An important function of this light is to determine, in a way, the 
lower limit of the tonal scale in our compositional structure, so far 
as it concerns exposure. At times, this light is used to establish a 
"photographic black" a phrase used to connote a dark tone, but 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 



261 




262 H. T. SOUTHER Vol 46, No. 4 

not one so definitely underexposed on the film that it assumes a 
muddy texture. 

The more intense the front-light is, the lower the contrast of the 
picture. We experience a decrease in dynamism but observe an in- 
crease in the pictorial quality. In addition, this important light es- 
tablishes the range of tonal gradations. It is very important in 
determining the elusive factor of quality in the lighting composition. 

The length of exposure of the negative also plays a great part in 
achieving a given tone to the image. By underexposing (shortening 
the exposure time of the subject) we accomplish a general reduction 
in tone and a compression of the tonal scale. The image becomes 
contrasted in nature. By overexposing the image becomes higher in 
over-all tone value and loses the quality of contrast. The laboratory 
may control this factor in printing, also. 

The name of the light front-light denotes its occupational 
position. It emanates from the region of the camera lens, frontally. 

The 3 lights mentioned above are the most important lights used 
in the construction of our compositional product. However, there 
are other lights which are variations of the above. They have func- 
tions of lesser importance, but certain conditions of shot construction 
find them necessary and invaluable. 

Three-Quarter Crosslight, or "Kicker." The "kicker" partially 
assumes the nature of the back-light. It "limns" the object as does 
the light from which it is derived, but with a much thicker edge. 
It lines one side of the object only, shown in Fig. 6. 

Sometimes two of these lights are used, one for each side, as seen 
in Fig. 7. It can be situated lower than the back-light. Because it 
is to one side, it may be moved over far enough to escape the camera 
line. 

The name of the light is significant. It literally "kicks" the object 
out of the background, and makes it seem dimension ally forward in 
the shot. 

Employment of Lighting Method. The thought now presents 
itself that the camera might be able to photograph the subject from 
any angle with a given lighting setup. This variation of shooting 
angle might affect the editorial composition, but pictorially, and from 
the standpoint of plastic interpretation, the subject should translate 
itself very satisfactorily to the screen. 

This conception involves the necessity for "taking the bull by the 
horns." Given a starting point for the beginning of the lighting proc- 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 263 

ess, it behooves us to light the first unit, set it definitely in its po- 
sition, adjust its intensity, and then proceed to work completely 
around the subject until we return to our starting point. 

Adhering to our basic precept of achieving a difference of intensity 
for every plane of the subject, we must at all times bear in mind the 
following points : 

(1) From our study of color we have found that all colors have a difference in 
actinic quality. Some colors require more light to bring them up to the required 
exposure level, some require less. 






LAMP B 




LAMP A 



POLISHED SURFACE 



FIG. 8. The effect of light reflection from an object to the camera with changing 

position. 

(2} The texture of the material of which the subject is composed may lend 
itself to a facile or a difficult transmission of light. Inasmuch as exposure depends 
on light transmission from the object, this factor is most important in determining 
the intensity with which we illuminate that particular area. 

(5) The angle of incidence of the light has a very decided bearing on the in- 
tensity with which it reaches the camera lens by reflection. If the angle of re- 
flection to the lens exactly equals the angle of incidence upon the subject, the 
transmission is much more effective. This means that the actual intensity of the 
light upon the subject must be gauged by sight from the lens position. Substances 
with sheen or polish, therefore, require much more light if these 2 angles do not 
coincide, and much less if they do. This is shown graphically in Fig. 8. 

Water is the most difficult of subjects to light. We may only 
photograph the halations and reflections on the water or, of course, 
those things which can be seen through the water. 

A mirror represents impossibility. We may photograph only 
those objects which are reflected into the mirror. 



264 



H. T. SOUTHER 



Vol 46, No. 4 



Portraiture. In connection with the art of motion pictures, it is 
well to realize that emotion in the mind of the audience is based on 
visual sensation. This sensation is a process that has for its founda- 
tion the material reality of the actual things, and the human beings, 
before the camera. The nearer these things are to the camera, the 
nearer are they to the audience. This closeness promotes a stimulus 
which is in direct ratio to the size of the image on the screen. 




1234 



FIG. 9. Diagram listing the planes of the face. 



Thus, a close-up of the face offers a prime means of communicating 
emotion to the audience. Through a near view of the subject, we may 
observe the generation of an idea, the reaction to a given occurrence, 
or the pictorial qualities of the idealized form. 

This last brings us face to face with one of the most important crea- 
tive problems of the cameraman. It consists in portraying in a close- 
up of the subject, either the realized form or the idealized form through 
the subtle mechanics of lighting. 

Idealization. We shall brook no argument with those who claim 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 265 

that photogenics engender a false or superficial standard of beauty, 
nor will we agree wholly with those who claim that the subject being 
photographed must be characterized by a socially or historically 
significant treatment (the realized form) . 

It follows that we cannot exclude the possibility of looking at a 
subject from a certain standpoint with a measure of objectivity even 
though the illusion of perfection has been the result of, perhaps, an 
aesthetic formula or recipe. It is perfection for which we strive, 
and this should result in pictures attaining true beauty. This 
ephemeral quality we achieve through scientific and realistic motiva- 
tion creatively applied to the camera. 

The basis of formal beauty lies in an involved ideological complex. 
For hundreds of years, since the time of the Greeks, the ideal frontal 
contour of the human head has been subconsciously accepted as be- 
ing egg shaped. 

It would seem extreme to say that if the head did not assume this 
egg shape, it could not be beautiful, but we must understand that the 
concept of facial perfection is not indigenous to a group of people, it 
is peculiar to the individual. It may not be easily fitted into the 
vise of uniformity. The monstrous lips of African women charm 
their men. Cleopatra by present standards is not what one would 
term attractive. The tyranny of fashion once embraced the "wasp 
waist' ' (1 7th century) . The Romans had their noses. 

In the pictorial synthesis of the face, the mind unconsciously takes 
those forms which deviate from the norm and attempts to make them 
conform in the mind's eye with this accepted ideal of the individual. 

The less effort entailed in making the subject conform, the more 
pleasurable is the emotional reaction of the audience. It is the 
cameraman's duty to promote this rationalizing process on the part 
of the viewer with every means at his command. These means are 
based on the phenomenon of optical illusion. 

The Planes of the Face. As seen from the frontal position, the 
face offers a series of 5 planes. If we interpret the face through the 
lighting treatment given these planes, shown in Fig. 9, we achieve 
the maximum in plastic rendering of the physiognomies. 

The Necessity for a Standard. Now, no doubt, there are as 
many conceptions of the ideal facial form as there are individuals to 
imagine them. But for purposes of illustration, we may not equivo- 
cate. We shall assume the Greek form as our standard of beauty. 
Given a starting point, we can be left to indulge ourselves in any 



266 



H. T/SOUTHER 



Vol 46, No. 4 



change, and we please our artistic side by preserving a subjective 
approach to beauty. 

Horizontal Divisions of the Face. This arbitrary facial contour, 
as set forth before, we shall assume to be egg shaped. Next we shall 
assume this form to be divided into 3 major equisections, seen in Fig. 
10. We then divide this form in hah 7 by a line. 

The intersection of the first sectional line is the place where the 
eyebrows would meet, if they continued. 



THREE 



MAJOR 



SECTIONS 




BONE STRUCTURE 

BONE STRUCTURE . 

EYEBROWS MEET IF PROJECTED 

CENTER WvEYES 

E NO Of NOS BONE STRUCTURE 

NOSE ENDING 

CENTER OF MOUTH 

BEGINNING OF BALL OF CHIN 

NARES EVEN WITH EYES 
UPS EVEN WITH EYES 
JAW EXTENDS OUTSIDE EGG 



FIG. 10. The divisions of the face. 



The intersection of the vertical line at the next dividing line de- 
termines the tip of the nose. 

Each of these 3 major divisions is subdivided into 3 minor ones. 
In the first case, the added horizontal lines determine the 2 ridges of 
the bone structure which compose the brow. 

The next division is more important. The first intersection hori- 
zontal line runs through the centers of the eyes. The second line 
intersects the end of the bone structure in the nose, and the beginning 
of the cartilaginous portion. 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 267 

The next line crosses the middle of the mouth. The last line es- 
tablishes the declivity which marks the beginning of the ball of the 
chin. 

Vertical Divisions of the Face. The vertical divisions of the ideal 
physiognomy are these : 

At that line where the eyes are placed, the lateral space is 4 eyes wide. The 
distance between the eyes is one eye in width. 

The corners of the mouth extend to a point directly underneath the pupil. 

The nares of the nostrils extend one-half this distance. 

The jaw extends slightly outside of the confines of the head contour. 

The hair areas assume an optional dimension. 

Formation of the Side of the Face. From the side of the face 
we observe that the ears are level with the eyes. 

The cheek is a flat plane (sometimes slightly protruding, sometimes 
slightly hollow) and extends almost up to the frontal region of the 
eye. 

The jaw line connects with the lobe of the ears. 

Application of Principles to the Idealization Process. The ap- 
plication of these principles to achieving the idealized form photo- 
graphically is, of course, dependent upon the manner in which the 
subject differs from our individual concept of the ideal. We classify 
below some of the more important deviations, and their correction 
through the manipulation of lighting. 

Corpulence. The presence of excess adipose in the face may be 
corrected in several ways : 

(1} Raising the Key -light. The key-light may be raised and thus increase the 
downward thrust of the shadows on the face. The effect is to compress the width 
of the subject seemingly through increasing its length. (Lines and angles pro- 
mote the illusion of length in the direction taken by them.) 

(2) Cross Key -lighting. Taking the key-light more around the subject helps 
very materially in this respect through our illusion phenomenon. The more cross 
positioned the key -light, the less area is illuminated from the camera position. 
(Dark is less noticeable with respect to area than light. Most of the face is in 
mezzotone.) 

Forehead Area Too Large. Subduing the light incident upon the 
brow apparently decreases its size. (Again dark is less pronounced 
than light.) 

Double Chin. Once more we have cause for a higher than nornial 



268 H. T. SOUTHER Vol 46, No. 4 

position of the key-light. This throws the underside of the chin into 
darkness. Less attention is directed to the fault. 

Nose Faults. Retrousse and beak noses require frontal composition 
in regard to camera position. Profile composition is fatal. 

The Eyes. The eyes, because of the active muscles surrounding 
them, are the most expressive feature of the face. In area, however, 
they do not possess the amount their importance would seem to de- 
mand in relation to other physiognomical elements. We may di- 
rect attention to the eyes in various ways so that they will appear to 
be as really significant as they are. Except for deformity, the eyes 
never seem too large. In fact, size characterizes their beauty. 

(1) Diminutive Eyes. Except for make-up, illusory lighting per- 
haps does more for the features in general and the eyes in particular, 
than movement, expression, or any other artifice employed by the 
individual. 

(2) The Eye-Light. A special light consisting of a small unit sit- 
uated near the camera lens will reflect back into the film as 2 pro- 
nounced points of hot light. The halation commands attention and 
is very pleasing in effect. 

(3) Focus. Sharp focus on the eyes is always important, but in 
portraiture, is absolutely essential. 

(4) Masking the Eye-Light. If the eye-light is masked off from the 
rest of the face and allowed to impinge only as an elongated rec- 
tangle horizontal with the plane of the eyes, the law of illusion will 
manifest itself in a definite enlargement of this area. A correlative 
result of this operation is to widen the head, subdue an otherwise 
unphotogenic facial composition, and direct the attention of the 
viewer to a specific reaction of the eyes alone. 

Photographing Blue Eyes. Blue eyes are very high in actinic 
value. The iris tends to photograph much lighter than would be 
ordinarily observed in everyday life. This results in a "washed out" 
and unexpressive appearance. The use of a magenta or red gelatin 
on the eye-light assists in photographing subjects with eyes of this 
type. These act as a filter and result in a much darker exposure. 
The result is a truer rendition of the subject characterization or, at 
least, control over the effect. 

Wrinkles. Wrinkles are the enemy of idealized photographic 
interpretation. The remedy lies in the employment of reduced con- 
trast. This requires frontal placement of the key-light, increase of 
fill-light, and the use of optical filters over the lens of the camera. 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 269 

DISCUSSION 

The Illusion of Depth in Motion Pictures 

QUESTION: The question is asked as to how the monocular method might be 
used. 

LT. SOUTHER: The paper as presented means to suggest that the "monocular 
method" is currently employed in the taking of motion pictures, but not to the 
fullest extent possible. The thought implied by present-day use of the term 
"stereoscopic perception" includes as a prerequisite for viewing the use of spec- 
tacles or a mechanical device to cause a "right-eye picture" to reach the right eye 
and a "left-eye picture" to reach the left eye, and it was the purpose of this paper 
to present for consideration the fact that binocular perception by the brain of two 
disparate views of an object is not altogether necessary in achieving a superior 
illusion over that commonly obtained today. 

The Fuller patents make use of the precepts exposed in this paper; that is, that 
three-dimensional perception is possible monocularly. Working on this premise, 
a dual image is imposed on the screen. By placing a line grating in front of the 
screen, the patent presumes to prevent the eyes or eye from scanning the image 
and thus duplicating the function which has already been performed by the cam- 
era. This line grating also presumes to prevent the eyes or eye from focusing on 
the image, thus duplicating another function already performed by the camera. 
Through the prevention of such duplication, the brain is allowed to perceive 
normally, through the eyes, the intent of the illusion originated by the camera. 

QUESTION : The impression was given that the line-grating process ordinarily 
used 2 cameras and twin projectors, and that the grating prevented the image of 
one projector from being seen by both eyes. 

LT. SOUTHER: That is an obvious method of purveying a right-eye image to 
the right eye, and vice versa. However, a complication is involved in that either 
movement of the head from side to side is necessary or, obviously, movement of 
the screen from side to side. The line grating as employed, with the intention 
propounded by the patent previously mentioned, serves a different function, 
namely that of preventing the eyes from duplicating actions already performed by 
the camera and thus spoiling the three-dimensional effect. This is rather an 
obtuse subject, I must admit. It requires considerable study and undoubtedly 
will receive a great deal of attention by experimenters in the future. I do feel, 
however, that the solution is not too far off, although it is not here at the present. 

QUESTION: Do we get the same effect in viewing a motion picture with one eye 
as we do with 2 eyes? 

LT. SOUTHER: I believe that the three-dimensional effect, observable through 
the line-grating method we have been discussing, would be increased not more 
than 50 per cent by viewing it with 2 eyes, providing the second eye was equally 
efficient as the first eye during the observation. In the reel that we have just 
seen, no special process was used other than an attempt to employ all the possible 
aids to roundness perception possible with current production methods. In the 
making of this reel everything was done to include as many foreground objects 
framing the picture as possible, in order to show one of the important aids to 
achieving illusion that we can use at present; that is, the relative apparent 
movement of objects when the camera is in motion. To answer your question 



270 H. T. SOUTHER Vol 46, No 4 

specifically, I do not think that there would be any vastly superior effect in 
viewing the usual motion picture with 2 eyes instead of one. 

QUESTION : I have observed, in viewing Grandeur film, a pronounced increase 
in three-dimensionality when it was being shown some years ago. Have you any 
idea why this was so? 

LT. SOUTHER: I believe that the increase in three-dimensional perception in 
this case was caused by the fact that the usual unnatural angle of the viewer was 
less pronounced because of the increased image size on the screen. We must 
realize that today, in viewing a motion picture with an aspect ratio of 3 to 4, we 
are more violently violating our visual continuity by confining the angle of view. 
We place ourselves in the theater in what would appear to be a long black tunnel, 
and view a scene transpiring in what would seem to be an opening at the end 
of this tunnel. Certainly this is something we are not used to in everyday life", 
and calls for a particular type of accommodation on the part of the viewer which 
must result in some irritation. I believe that this irritation reduces, uncon- 
sciously, the viewer's ability to perceive solidity. I would say that in my opinion 
the larger Grandeur screen, because it does not limit the angle of view so much, 
gives a greater three-dimensional effect, but only for that reason. 

QUESTION : Could we have that reel run again? I would like to ask the mem- 
bers to confine their observation to the screen with one eye this time. I would 
also suggest that the members make a viewing tube out of their hands in order to 
exclude observation of the rest of the room. The purpose of this is to see if an 
increase in the three-dimensional effect is caused by such viewing. 

[Ed. Note. The film was re-run, and a show of hands was called for to indicate 
whether an increase in solidity had resulted from this observation with one eye and 
through the exclusion of the room in the field of view through the viewing "hole" 
formed by the hand. The show of hands indicated that approximately 80 per cent 
of the members present had experienced an increase in the feeling of three-dimen- 
sionality.] 

QUESTION : Can we have the film run once more, and this time ask the mem- 
,bers to cover one eye only and view the film without the tube formed by their 
hand? 

[Ed. Note. When the film was run, a show of hands was called for as to whether 
the effect was more pronounced or less pronounced than in the last test. Ap- 
proximately 75 per cent of the members believed that there was less effect than 
before. The film was then run once more upon request, and viewed again with 
both eyes.] 

The Theory and Practice of Lighting for the Camera 

QUESTION : How is the intensity of the key-light controlled on the actor's face, 
particularly in a "two-shot?" 

LT. SOUTHER: If I understand your question properly, I would say that the 
intensity of the key -light on an actor's face is determined by tests before the start 
of the production. The reflectiveness of certain make-ups differs, and the addi- 
tion or decrease of key-light for a certain make-up must be predetermined. For 
all normal scenes thereafter throughout the picture the key-light intensity is 
adjusted to this same tested level. This presupposes the use of a controlled 



April, 1946 THE THEORY OF LIGHTING FOR THE CAMERA 271 

standardized development process, such as that employed by Twentieth Century- 
Fox. It is not unusual to have an entire day's shooting print on a single light set- 
ting. Where two actors, both using make-ups of different reflectiveness, are 
illuminated by a single key-light, scrims must be employed on that side of the lamp 
to adjust the particular make-up to the degree of light transmission determined 
upon by tests before the start of the production. 

It was pointed out during the demonstration that the use of dimmers on in- 
candescent units calls for judgment in the choice of the power of the illumination 
unit. A unit too large, dimmed to the required intensity for a close subject, re- 
sults in illumination of poor spectral quality, which in printing causes muddy 
texture owing to its low actinic value. The proper regulation of arcs, when used 
as key-lights, calls for extreme care in placement of the illuminating unit and 
the precise application of scrims to control intensity. Dimmers are not currently 
practical on arcs. 

[Ed. Note. The set shown in Fig. 1 was duplicated at the Conference with 
model for purposes of demonstration. The units were lighted one by one, and 
their names and purposes explained.] 



WESTREX STANDARD SOUND FILM REPRODUCER* 

G. S. APPELGATE** AND J. C. DAVIDSON f 

Summary. A simplified sound film theater reproducer is described. The 
machine holds the film under constant tension and has a minimum of moving parts. 

A review of the theater field for sound film reproducers has indi- 
cated a definite need for a machine that will fulfill the requirements of 
the relatively large field where equipment cost is a prime factor. 
Since the requirements permit no compromise on quality or perform- 
ance, the need for a new basic design was indicated. Accordingly, 
the development of a machine was undertaken on this basis and each 
part was given a careful scrutiny to justify its use from the viewpoints 
of either assured performance or operating necessity. This work has 
resulted in a reproducer whose component parts are surprisingly few 
and yet whose film propulsion performance falls well within the 
recommended standards of the Research Council of the Academy of 
Motion Picture Arts and Sciences. 

A front view of the machine is given in Fig. 1, showing the film, 
lamp, and photocell compartments. The film path is perhaps as 
simple as has been offered to the theater industry and approaches a 
straight vertical line from the holdback sprocket in the projector head 
to the lower magazine with no free loops to be set. The film is under 
a tension of approximately 300 grams from the lower sprocket in the 
projector to the sound sprocket in the reproducer. 

The threading operation is both quick and simple. The film is 
passed under the upper filter arm roller, over the scanner drum, and 
over the lower filter arm roller. Before engaging the sound sprocket 
the film is pulled down until the arrow and line on the filter arm 
assembly are in approximate registry. The film is then engaged 
around and under the sound sprocket and over the idler roller which 
guides it into the fire trap of the lower film magazine. 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Westrex Corporation, New York. 

f Electrical Research Products Division, Western Electric Company, Holly- 
wood. 
272 



WESTREX SOUND FILM REPRODUCER 



273 



The film filtering mechanism consists of a solid flywheel mounted 
on the scanner shaft which supplies the inertia element, 2 filter arms 
and rollers connected together by a spring, which furnish the compli- 
ance element where rates of flutter below 10 cycles are concerned, and 
viscous dashpot connected to the lower arm, which provides the 
damping element. Above 10 cycles the attenuation of disturbing 
frequencies by the film compliance gradually exceeds that furnished 
by the spring. While the filter parts are simple, an explanation of 
how they function involves a rather complicated electrical analogy 
and is covered in detail elsewhere. 1 It is sufficient here to state that 




FIG. 1. Front view of Westrex standard sound film reproducer. 



the upper filter arm, which is not directly damped by the dashpot, 
supplies an attenuation of the order of 12 db per octave against dis- 
turbances arising in the projector head. This permits the elimination 
of the free loop between picture and sound head with a reasonable 
factor of safety. A further factor of safety against slow rates of dis- 
turbances occurring in either the projector head or take-up magazine 
has been provided by the use of a new sprocket tooth with a relatively 
wide base. 

In the present film path where the film operates between 2 sprockets 
under a relatively fixed tension, it is desirable that the tension beyond 
the 2 sprockets be maintained at either a greater or lesser value than 
that existing between the sprockets. The tension of the film on the 



274 G. S. APPELGATE AND J. C. DAVIDSON Vol 46, No. 4 

incoming side of the upper sprocket, which is in the projector, is sub- 
stantially zero, so that the requirement is met at this point. 

The film leaving the lower, or sound sprocket, is at substantially the 
same tension as the take-up. As long as the take-up tension is main- 
tained at a higher value than that in the filtered path, the holdback 
sprocket can be eliminated with reasonable safety. In the event the 
take-up tension approaches the same value as that in the filtered film 
path the variations in take-up tension may cause the film to travel be- 
tween the limits of free play between the tooth and sprocket hole. 
The effect is, of course, attenuated by the belt action of the film on the 
sprocket and on the lower idler roller. 

To minimize the amount of disturbance that might occur from this 
source, a new sprocket tooth has been developed which more nearly 
fits the sprocket hole, with due consideration for film shrinkage, etc. 

The scanner assembly consists of a scanner drum, shaft, and fly- 
wheel and mounts in the main frame as a unit. The filter arm 
assembly also mounts as a unit and consists of 2 arms and rollers 
mounted on cone pivots operating in ball races to produce a minimum 
of bearing friction and lateral play. The upper roller is provided 
with a lateral adjustment to provide means for aligning the sound 
track with respect to the principal axis of the optical system. The 
dashpot is easily removed for inspection. The liquid used in the 
dashpot has been chosen for its small viscosity change with tempera- 
ture to insure a sufficiently uniform damping characteristic. 

The optical system consists of a flexibly mounted lamp bracket 
which takes the theater prefocused base 7V2-amp, 10-v exciter lamp, 
the Bausch and Lomb 41-87-35 objective, a collective lens, and a 
photocell. The collective lens images the aperture of the objective 
lens onto the cathode of the photocell and gives approximately a uni- 
form area of variable-intensity illumination on the cathode with 
modulation of the light beam. 

The photocell is flexibly mounted and is coaxially coupled to a 
remotely located photocell amplifier. The photocell mounting plate 
is so designed that it is interchangeable with a 2-stage photocell 
amplifier, should that arrangement be preferred to the coaxial cable 
coupling. 

The drive arrangement is somewhat different from current practice 
and contains some interesting features, particularly when considered 
from the standpoint of ease of assembly and alignment. It is shown 
in Fig. 2. With the exception of the scanner assembly, the sound 



April, 1946 



WESTREX SOUND FILM REPRODUCER 



275 



sprocket shaft is the only one that is mounted in the main frame cast- 
ing. This shaft is driven by the motor through double vee-belts. 
Since the motor is flexibly mounted, the vee-belts provide a coupling 
that transmits very little motor vibration to the film drive mechanism. 
Double belts are used to minimize the effect of belt irregularities on 
the machine's performance. The belt drive also simplifies the motor 
alignment problem. Different motor speeds are accommodated by 
a change of pulleys. If synchronous interlock operation is desired, 
the vee-belts are replaced by a silent chain drive. 




FIG. 2. Drive gear assembly. 

Also mounted on the sprocket shaft are the silent chain sprocket 
for driving the projector head and a gear to couple with the take-up 
pulley shaft. The take-up pulley and its driven gear are located by 
means of a stub shaft on a spider that centers on the sprocket shaft 
and is locked by 2 screws to bosses on the main frame. An idler to 
tension the projector drive chain is also mounted on the spider. The 
take-up belt tension is adjusted by loosening 2 screws and rotating the 
spider. 

Fig. 3 shows a chart of flutter as measured on a preproduction 
model of the reproducer. It will be observed that the total integrated 
flutter from 2-200 cycles does not exceed =*= 0.09 per cent, while no dis- 



276 



G. S. APPELGATE AND J. C. DAVIDSON Vol 46, No. 4 



crete disturbance between 2 l / 2 and 200 cycles is greater than 0.05 
per cent. Occasional irregular disturbances in the bands between 0-1 
and 1-2 V 2 cycles reach as high as 0.06 and 0.08 per cent, respec- 
tively. This results from the fact that full advantage of the new 
sprocket tooth could not be taken because a few laboratories still 
release on film that does not have standard positive perforations. 

The type of film path and filter system used in this reproducer is 
substantially identical with that used in several designs of studio re- 
corders and rerecorders. Tests set up to simulate operation under 



.15 

.oe 

.06 
.04 

< 






PRE 


>EN 


r 


AC 


A)l 


M" STAND/ 


RD F 


OR 


TO 


T/ 


L 




_ JTTER 




































-- 


























"X 






























































.X 












TOTAL F 


JUTT 


ER 


2- 


2C 


'0 


ny 














































^ 


MM 




































































. 






























~1 
























r 




























































J 1 10 


100 2C 



FLUTTER FREQUENCY IN CYCLES PER SECOND 

FIG. 3. Flutter performance chart. 



adverse conditions indicated that little or no effect was discernible in 
the reproduced sound. 

Film weave in this system appears to be negligible. In one test, 
sprocket eccentricity to the extent of 18 mils was introduced and while 
the filter arms were set into considerable motion, only a very small 
increase in measured flutter was found. A piano recording was re- 
produced under this condition and the listeners were unable to detect 
the effect of the sprocket eccentricity. The photocell network and 
coaxial cable pick up no machine noise that is audible under any 
normal condition of operation. 

In conclusion, it is felt that the performance as Veil as the sim- 
plicity of this reproducer have fully met the design expectancy. 



April, 1946 WESTREX SOUND FlLM REPRODUCER 277 

The authors desire to take this opportunity to acknowledge the 
contributions to this design made by the engineering department of 
the Century Projector Corporation. 

REFERENCE 

1 DAVIS, C. C. "An Improved Film Drive Filter Mechanism," to be published 
in a forthcoming issue of /. Soc. Mot. Pict. Eng. 



WESTREX MASTER SOUND FILM REPRODUCER 5 
G. S. APPELGATE** AND J. C. DAVIDSONf 



Summary. The following article describes a new design of film reproducer 
which has brought the flutter content to a minimum, and which will maintain its 
low flutter rate regardless of ordinary wear and tear. Other unique features embodied 
in the design are a new optical system, rugged construction and ease of maintenance. 



Over 7 years have elapsed since the Western Electric Company de- 
veloped its last theater sound film reproducer. The interval of time 
has been sufficiently long to permit a well-considered evaluation of 
how much has been accomplished as well as to point the direction that 
future development should take. 

When the former machines were in good adjustment and the film 
was in good condition, the performance obtained appeared to meet the 
needs of the theater industry rather satisfactorily. The machines, 
however, were subject to two vagaries which sometimes affected their 
performance to a point that was discernible to a critical listener. 
These may be summed up as scanner-bearing trouble and the physical 
condition of the film. Either resulted in an increase in the low rates 
of flutter. 

It seemed, therefore, that future endeavor should point first toward 
the attainment of assured stability of operation and then toward such 
further improvement in performance as may seem to be justified. 

In the new reproducer a film path and filter system have been de- 
veloped wherein the film compliance is utilized to obtain attenuation 
of high-frequency disturbances while at the same time means are 
provided to minimize the possibility of the vagaries of the film from 
setting up low rates of flutter sometimes referred to as "wows." 
The theoretical basis for the filter system is discussed in detail else- 
where. 1 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Westrex Corporation, New York. 

t Electrical Research Products Division, Western Electric Company, Holly- 
wood. 
278 



WESTREX MASTER SOUND FILM REPRODUCER 



279 



A front view of the machine is given in Fig. 1, with the doors open 
and with the photocell amplifier cover plate removed. Film has been 
threaded to show the film path. 

Film Path. The film is received from the projector head as a 
free loop and is held in contact with the scanning drum by a pressure 
roller. It leaves the drum under controlled tension, passes over the 
damped spring compliance roller, and is then engaged by the sound 
sprocket. There is a free film loop between the sound sprocket and 




FIG. 1. Front view of Westrex master sound film reproducer. 



the holdback sprocket. Thus it is seen that the film path between 
the scanner and the sound sprocket is isolated from disturbances 
originating in either the picture head or take-up magazine by free 
loops. The pressure roller is provided with means for lateral move- 
ment to align the sound track with the optical axis. This adjustment 
is provided with an indexed head so the correction for a misplaced 
sound track can be preset and the alignment can be returned to 
normal without requiring special attention. 

With the machine running it is desired that the film between the 
scanner drum and the sound sprocket be under a tension of approxi- 



280 



G. S. APPELGATE AND J. C. DAVIDSON Vol 46, No. 4 



mately 300 grams. Since there is a free loop above the scanner 
drum, this requires that a constant load be applied to the latter. 
This is accomplished by associating an eddy current drag with the 
scanner flywheel. The eddy current drag is obtained by a copper 
ring rotating in a magnetic field produced by permanent magnets. 
The amount of drag is controlled by altering the position of the mag- 
nets and it, in turn, determines the amount of tension in the film be- 
tween the scanner drum and 
the sound sprocket. 

Scanner Assembly. The 
scanner assembly consists of 
a scanner drum, a solid fly- 
wheel, and an eddy current 
ring and permanent magnets 
mounted in a closed casting. 
It mounts as a complete as- 
sembly in the main frame wall 
by means of 3 cap screws. Its 
radial position is determined 
by a dowel pin, thus facilitat- 
ing its removal and replace- 
ment without adjustment 
change. In the previous 
scanner designs irregular bear- 
ing friction has been a major 
source of trouble, particularly 
since the film was under rela- 
tively light tension. In the 
present design, the scanner 
FIG. 2. Drive gear assembly. shaft is supported in 2 small 

3 /8-in. outboard ball bearings 

and as a result, disturbances from the scanner bearings have been 
practically eliminated. 

Compliance Damper Assembly. The compliance damper assem- 
bly consists of a casting in which is mounted a pivoted arm. The 
arm is spring tensioned and has a viscous damper attached to it. 
The arm supports a ball-bearing mounted roller over which the film 
rides. The arm pivots have been designed to have a minimum of 
friction and lateral play and consist of cone points operating in ball 
races. The spring tension is adjustable by means of a locknut and 




April, 1946 WESTREX MASTER SOUND FILM REPRODUCER 



281 



screw. The cup containing the damping fluid is readily removed for 
inspection by loosening a thumbscrew. The damping fluid has been 
chosen to have a minimum of viscosity change with temperature. 

Optical System. The optical system consists of the prefocused 
base, 9-v, 4-amp exciter lamp, the Bausch and Lomb 41-87-35 objec- 
tive, a pair of collective lenses, and a photocell. The lamp assembly 
is flexibly mounted and contains the usual lamp and an auxiliary lamp 
with means for adjustment of the filament in the vertical plane. In 
case of an emergency burn-out, the auxiliary lamp can readily be 



t.15 
u 

^ 

.0 
.04 
+ 02 

J 






PRE 


SEr> 


T 






Mr STAND 


tRD 


'OR 


T( 


T 


U. I 


.UTTER 




















































V 






1 1 


-- 





- 







. , 

























































TOTAL 


UJT 


FER 


2- 


2( 


K>" 








<T 






















' 












































































r 


































I" 


1 












1 


10 


100 21 



FLUTTER FREQUENCY IN CYCLES PER SECOND 

FIG. 3. Flutter performance chart. 

moved into its correct position on the optical axis, and power trans- 
ferred to it by operating the auxiliary lamp switch. The Bausch and 
Lomb lens tube has been widely used in the theater field and needs no 
description. 

Between the film plane and the photocell is a doublet lens followed 
by a single lens. The doublet images the film plane in the aperture 
of the single lens, thus providing a plane in which separator lenses 
may be placed for scanning 100-mil push-pull or double track should 
this requirement arise. The single lens images the aperture of the 
doublet onto the cathode of the photocell. Under this condition the 
photocell cathode sees a spot of light of uniform area and variable 
in intensity, irrespective of the nature of the modulation on the film. 



282 G. S. APPELGATE AND. J. C. DAVIDSON Vol 46, No. 4 

The collective lens system assembly is removable by means of a 
thumbscrew for cleaning. 

Drive Mechanism. Fig. 2 shows a rear view of the machine 
with the rear cover and motor cover removed. It shows the drive 
gear assembly. 

The drive mechanism has a vertically mounted motor that is 
flexibly coupled to a vertical shaft. This shaft drives a horizontal 
shaft through a pair of right-angle helical gears. The horizontal 
shaft contains 2 helical drive gears which drive 2 cross shafts. One 
cross shaft supports the sound sprocket. The second supports the 
holdback sprocket and the chain sprocket for driving projector heads 
other than the Century head. A third cross shaft provides the drive 
for the take-up and is gear coupled to the holdback sprocket shaft. 

When the sound head is set up with a Century projector a new type 
of coupling has been provided. This consists of a second vertical 
shaft in line with and coupled to the vertical shaft in the projector 
head by a flexible shaft. This eliminates the relatively expensive 
chain or pinion drive in current use with other types of projector 
heads. 

A flywheel is provided on the motor shaft to insure a sufficiently 
slow starting time and a hand brake is available for a quick stop in 
case of a film break. 

Transmission System. The photocell and a 2-stage photocell 
amplifier are assembled on a small flexibly mounted chassis which is 
removable for inspection. Two Western Electric 6AK5 tubes are 
used. The first stage is a gain stage, while the second is an impedance 
transforming stage of the cathode follower type. A separate gain 
attenuator is mounted on each machine. The attenuator is located 
beyond the photocell amplifier at a point of relatively high level. 

Performance. Fig. 3 shows a chart of the performance from the 
flutter standpoint as measured on a preproduction model of the re- 
producer. It will be noted that at no flutter rate does the amount of 
flutter exceed == 0.04 per cent, while the total integrated flutter from 2 
to 200 cycles does not exceed 0.08 per cent. This performance is 
considerably better than the flutter requirements of the present 
standard of the Research Council of the Academy of Motion Picture 
Arts and Sciences. The time of recovery after the passage of a film 
splice is less than one second. There is little tendency for the film to 
weave in this tensioned system. Sufficient studies have been made 
to indicate that this performance will be realized in commercial 



April, 1946 WESTREX MASTER SOUND FlLM REPRODUCER 283 

production and will be maintained in the field with a minimum of 
service. 

The transmission system has been sufficiently isolated so that no 
machine noise can be heard above photocell hiss with the machine 
running, the lamp on, and no film in the optical path. 

The authors wish to take this opportunity to acknowledge the 
numerous contributions of machine design that have been made by 
the engineering department of the Century Projector Corporation. 

REFERENCE 

1 DAVIS, C. C. : "An Improved Film- Drive Filter Mechanism," to be published 
in a forthcoming issue of /. Soc. Mot. Pict. Eng. 



AMERICAN STANDARDS ON MOTION PICTURES 



FOREWORD 

Industrial standards are by nature not static rigid limits that once 
established must be adhered to religiously for all time. They are 
subject to change as are the changing times and so must be reviewed, 
revised, and brought up to date periodically, if the industry that pre- 
pared them is to continue to look on them with favor. 

Once each year, Sectional Committee on Motion Pictures Z22 of 
the American Standards Association reviews all American Standards 
on Motion Pictures, determining whether the substance and form of 
presentation are still suitable to continue serving their original pur- 
pose. 

Following the October 17, 1945, meeting of Committee Z22, eight 
American Standards were reaffirmed, subsequently approved by the 
ASA and will remain active for the present time. 

The war procedure of the ASA requires that all American War 
Standards shall be reviewed after the end of hostilities to determine 
whether they should be : 

(1) Reapproved and continued as American Standards. 

(2) Revised, if necessary; or 

(5) Dropped if they have outlived their usefulness . 

Consequently all published American War Standards on Motion 
Pictures were also reviewed. Fifteen were approved to continue now 
as American Standards. They have been assigned Z22 numbers and 
are published here in American Standard format. 

Three of the reaffirmed Z22 Standards, Z22.37, Z22.,38, Z22.39, 
were published previously in the August, 1944 JOURNAL, and since 
these are of comparatively recent date they are not reprinted here. 

The published American Standards are the result of effective col- 
laboration over a long period by a large number of important groups 
in the motion picture industry. The continued and valued co-opera- 
tion of the many individual members of the Armed Forces, the Re- 
284 



AMERICAN STANDARDS ON MOTION PICTURES 



285 



search Council of the Academy of Motion Picture Arts and Sciences, 
and the Society of Motion Picture Engineers is gratefully acknowl- 
edged. 

With the exception of the 3 Standards appearing in the August, 
1944 JOURNAL, the publication of the 20 Standards listed here super- 
sedes all previous publication of these American Standards on Motion 
Pictures. 

AMERICAN STANDARDS 

Z22.2-1946 35-Mm Sound Film; Emulsion and Sound Record Positions in 

Camera Negative 
Z22. 3-1946 35-Mm Sound Film; Emulsion and Sound Record Positions in 

Projector Positive (for Direct Front Projection) 

Z22.9-1946 16-Mm Silent Film; Emulsion Position in Camera Negative 
Z22. 15-1946 16-Mm Sound Film; Emulsion and Sound Record Positions in 

Camera Negative 

Z22. 21-1946 8-Mm Silent Film; Emulsion Position in Camera Negative 
Z22.40-1946 Sound Records and Scanning Area of 35-Mm Sound Motion 

Picture Prints 
Z22. 41-1946 Sound Records and Scanning Area of 16-Mm Sound Motion 

Picture Prints 
Z22. 42-1946 Sound-Focusing Test Films for 16-Mm Sound Motion Picture 

Projection Equipment, Specification for 
Z22. 43-1946 3000-Cycle Flutter Test Film for 16-Mm Sound Motion Picture 

Projectors, Specification for 
Z22. 44-1946 Multifrequency Test Film Used for Field Testing 16-Mm Sound 

Motion Picture Projection Equipment, Specification for 
Z22. 45-1946 400-Cycle Signal Level Test Film for 16-Mm Sound Motion 

Picture Projection Equipment, Specification for 

Z22. 46-1946 16-Mm Positive Aperture Dimensions and Image Size for Posi- 
tive Prints Made from 35-Mm Negatives 
Z22. 47-1946 Negative Aperture Dimensions and Image Size for 16-Mm 

Duplicate Negatives Made from 35-Mm Positive Prints 
Z22. 48-1946 Printer Aperture Dimensions for Contact Printing 16-Mm 

Positive Prints from 16-Mm Negatives 
Z22. 49-1946 Printer Aperture Dimensions for Reversal and Color Reversal 

Duplicate Prints 

Z22 . 50-1946 Reel Spindles for 16-Mm Motion Picture Projectors 
Z22. 51-1946 Intermodulation Tests on Variable- Density 16-Mm Sound 

Motion Picture Prints, Method of Making 
Z22 . 52-1946 Cross-Modulation Tests on Variable-Area 16-Mm Sound Motion 

Picture Prints, Method of Making 
Z22. 53-1946 Resolving Power of 16-Mm Motion Picture Projector Lenses, 

Method of Determining 
Z22. 54-1946 Freedom from Travel Ghost in 16-Mm Sound Motion Picture 

Projectors, Method of Determining 



AMERICAN STANDARDS ON MOTION PICTURES Vol 46, No. 4 

One other group of 25 American Standards has been referred back 
to the SMPE for disposition. Two American Standards have been 
referred back to the Research Council of the Academy of Motion 
Picture Arts and Sciences, and 3 American Standards are being re- 
vised by a subcommittee of Z22. These 30 Standards will also be 
published in the JOURNAL as soon as they are formally approved 
through the procedure of the Research Council of the Academy of 
Motion Picture Arts and Sciences, the Society of Motion Picture 
Engineers, and the American Standards Association. 



April, 1946 



AMERICAN STANDARDS ON.MOTION PICTURES 



287 



American Standard 

Emulsion and Sound Record Positions 
in Camera for 35-Millimeter 
Sound Motion Picture Film* 


Reg. V S. Pat. Off 

Z22.2-1946 
First Edition: 
Z22.2-1941 




^ " 

CD 


^ 

^ 

^ 
^ 


a 


GUIDED EDGE 


\M 
VEL 


^^L, 




a 
a 
a 
a 


CD 

a 
a 
a 


^^^LIGHT BE> 
TRA 


X 


^jr 

CD 

in 1 1 1 


a 

CD 

a 

CD 






a 3 1111 
a 
a ,* 


\ 


a 
a 
a 





Drawing shows film as seen from inside f/i camera looking 
toward the camera lens. 



1. Emulsion Position 

1.1 The emulsion position in the camera shall be toward the lens, except for 
special processes. 

2. Speed of Projection 

2.1 The speed of projection shall be 24 frames per second. 



3. Distance Between Picture and Sound 

3.1 The distance between the center of the picture and the corresponding 
sound shall be 20 frames. 



*The title of this standard is the only revision from the 1941 edition. 



288 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard 

Emulsion and Sound Record Positions 

in Projector for 35-Millimeter 

Soun.d Motion Picture Film* 



Keg. V. S. Pal. Off. 

Z22.3-1946 

First Edition: 
Z22.3-1941 




LIGHT BEAM 



Drawing shows film as seen from the light-source in the pro/ecfor. 



1. Emulsion Position 

1.1 The emulsion position in the projector shall be toward the light-source, 
except for special processes. 

2. Speed of Projection 

2.1 The speed of projection shall be 24 frames per second. 

3. Distance Between Picture and Sound 

3.1 The distance between the center of the picture and the corresponding 
sound shall be 20 frames. 



*The title of this standard is the only revision from the 1941 edition. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



289 



American Standard 

Emulsion Position in Camera for 
16-Millimeter Silent Motion Picture Film* 



Reg. V. S. />. Off. 

Z22.9-1446 

First Edition 
Z22.9-1941 




LIGHT BEAM 
TRAVEL 



Drawing shows film as seen from inside the camera looking 
toward the camera lens. 



1. Emulsion Position 

1.1 The emulsion position in the camera shall be toward the lens, except for 
special processes. 



2. Normal Speed of Exposure 

2.1 The normal speed of exposure shall be 16 frames per second. 



*The title of this standard is the only revision from the 1941 edition. 



290 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 



American Standard 

Emulsion and Sound Record Positions in Camera 
For 16-Millimeter Sound Motion Picture Film* 



K,-K. V. S. I'ai. Off. 

Z2 2. 15- 1946 
First Edition: 
Z22.15-1941 




Drawing shows film as seen from inside the camera looking toward the 
camera lens. 



1. Emulsion Position 

1.1 The emulsion position in the camera shall be toward the lens, except for 
special processes. 

2. Speed of Projection 

2.1 The speed of projection shall be 24 frames per second. 






3. Distance Between Picture and Sound 

3.1 The distance between the center of the picture and the corresponding 
sound shall be 26 frames. 



*The title of this standard is the only revision from the 1941 edition. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



291 



American Standard 

Emulsion Position in Camera for 
8-Millimeter Silent Motion Picture Film* 



dee. V. S. Pat. Off. 

Z2 2.21 -1946 
First Edition: 
Z22.21-1941 



LIGHT BEAM 




Drawing shows film from inside the camera, looking toward the camera lens. 



1. Emulsion Position 

1.1 The emulsion position in the camera shall be toward the lens, except for 
special processes. 



2. Normal Speed of Exposure 

2.1 The normal speed of exposure shall be 16 frames per second. 



*The title of this standard is the only revision from the 1941 edition. 



292 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard 

Sound Records and Scanning Area 

of 35-Millimeter Sound Motion Picture Prints 



ASA 



Rrf. V. S. Pal. Off. 

Z22.40-1946 



AREA PRINTED 
IN SOUND PRINTER 

SHADED AREA TO BE EF 
OPAQUE ON PRINT 

VARIABLE AREA 
AND MATTED 
VARIABLE DENSITY 
RECORDS 

SHADED AREA TO BC EF 
OPAQUE ON PRINT 

FULL WIDTH 
VARIABLE DENSITY 
RECORD 

AREA SCANNED B 
REPRODUCER 




O 


n 


\ 





* UUIULU LUUL 

O.I92+-0- 001 IN. 


OUTER EDGE OF 


PRINTED AREA \, 
INNER EDGE OF 


jE 


4.68 +0.02 MM 


PRINTED AREA 


7.75 0-05 MM 
0.076+0001 IN. 


! 




TEC 


i 






m^~r 




O 

O 




\0 

|a 


WIDTH OF 
SOUND RECORD * 


MATTED V. D. 
(FULL WIDTH) AND j 
IOO*>MODULATIONVA 


;o 
p 

tf~L 


1.93+ 0.02 MM 
0.243*0.002 | N . 




4 

TEC 


. SOUND RECORD -1 

FlVELY ' 




' 6.17 0.05 MM 

o.ioot:o? IN. 




a-~ 




O 
O 

n 




io 

!o 


WIDTH OF 


SOUND RECORD* 


jo 
jo 
n 


2.54+O o . fpM M 
0.2430-OOI| N . 




___ : 


f 


SCANNED AREA 
WIDTH OF 


L 


6. 17+ 0.02 MM 
0.084 CXOOI IN. 


SCANNEDAREA ^" 2.13 + 0-02 MM 
HEIGHT OF I 0.001 2 + -aoooi 2 IN. 


SCANNED AREA 


0.030 +000! MM 






Distance Between Sound and Corresponding Picture The sound shall pre- 
cede the center of the corresponding picture by a distance of 20 it Vz frames. 



These Dimensions and Locations Are Shown Relative to Unshrunk Raw Stock 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



293 



American Standard 

Sound Records and Scanning Area 

of 1 6-Millimeter Sound Motion Picture Prints 



Reg. V. S. Pat. Off. 

Z2 2,41 -1946 



GUIDED EDGE 

0.013 IN. MAX 


3( 


a 
a 


X{ OUTER EDGE OF 


0.457 MM MAX 
O.I 10 IN. MAX, 0.098 IN. MIN 


S PRINTED AREA 
V 
^ DINNER EDGE OF 


2.794 MM MAX, 2.489 MM MIN 


\X PRINTED AREA 

\ 




0.058-2 IN. 


SOUND RECORD 




.473- 051 MM 


* o.oeolg;^t IN. 




a 
a 


* WIDTH OF 
1 SOUND RECORD 


'524!g;iii MM 


ij V.D. SQUEEZE OR 
lOO^o MODULATION 
V.A. 


. 


0.058* a002 IN. 


-* SOUND RECORD 


I. 473 0.05I MM 


0.080 - 001 IN. 




a 
a 


WIDTH OF 


2 032 0.025 MM 


SOUND RECORD 




0.058 a 02 IN. 


* SOUND RECORD 




|.473- 051 MM 


0.072 IN. MAX, 0.070 IN. MIN 




a 




WIDTH OF 


1.829 MM MAX, 1.778 MM MIN ^ 


SCANNED AREA 


! 


0.058 - 001 IN. 


SCANNED AREA 


,. 473 0.025 MM * 



PRINTED AREA 



VARIABLE AREA 

AND 

VARIABLE DENSITY 
SQUEEZE RECORDS 



FULL WIDTH 

VARIABLE DENSITY 

RECORD 



SCANNED AREA 



These Dimensions and Locations Are Shown Relative to Unshrunk Raw Stock 

"This dimension for the width of the sound record of variable density squeeze tracks and 
of variable area tracks at 100 percent modulation is based on present day equipment 
design. It is recommended that all future equipment be designed for a record width of 
0.060 0.001 inch. It is also recommended that existing equipment be modified tp pro- 
duce prints having variable density squeeze and 100 percent modulation variable area 
records with a width as close as practicable to 0.060 0.001 inch. 



294 AMERICAN STANDARDS ON MOTION PICTURES Vol 46, No. 



American Standard Specifications for 

Z22.42-1946 



Sound-Focusing Test Films *.*.*.*.<. 



for 16-Millimeter Sound Motion Picture Projection Equipment 



1. Scope and Purpose 

1.1 This specification describes test films to be used for checking the focus of the scanning 
beam of 16-mm sound motion picture projectors. 

2. Test Films 

2.1 The test films shall have an originally recorded vdriable-density sound track, heavily 
overmodulated and developed to high contrast so that the resultant track is essentially a 
square-wave track. 

2.1.1 The test films shall be of 2 types: 

Type A 7000-cycle recording for manufacturing and precision adjustment of 
sound focusing; 

Type B 5000-cycle recording for quick field adjustment of focusing. 

2.2 The sound track shall have correct azimuth within 5 minutes of arc. 

2.3 Each test film shall be provided with a suitable leader, title, and trailer. 

2.4 The standard length of the test films shall be 100 feet. 



NOTE: A test film in accordance with this standard is available from the Satiety of 
Motion Picture Engineers. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



295 



American Standard Specifications for 

3000-Cycle FlutterTestFilm 

for 1 6-Millimeter Sound Motion Picture Projectors 



Reg. V. S. Pat. Off. 

Z22.43-1946 



1. Scope and Purpose 

1.1 This specification describes a 3000-cycle 
sound test film for use in determining the 
presence of flutter in 16-mm sound motion 
picture projectors. 

2. Test Film 

2.1 Recording. The test film shall have 
either an originally recorded, direct-play- 
back positive variable-area sound track or 
an originally recorded variable-density 
sound track developed as a toe record. The 
recorded frequency shall be within 25 
cycles of the nominal 3000-cycle frequency. 
The modulation of the recording shall be 80 
=b 5 percent. The output level of the film shall 
be constant within it !4 db. (This is equiva- 
lent to an amplitude tolerance of 0.0015 
inch when recording variable-area sound 
track with a nominal amplitude of 0.055 
inch.) The recording shall be accomplished 
in a recorder so constructed as to keep the 
flutter content to the absolute minimum con- 
sistent with the state of the art. The total rms 
flutter content of the film shall be less than 
0.1 percent upon shipment by the test film 
manufacturer. The wave-form distortion of 
the recording shall not exceed 5 percent. 

2.2 Film Stock. The film stock used for the 
test film shall be cut and perforated in ac- 
cordance with the American Standard 16- 
Mm Sound Motion Picture Film; Cutting and 
Perforating for Negative and Positive Raw 
Stock, Z22.12-1941, or latest revision thereof. 



2.2.1 Resistance to Shrinkage. The 

film stock used for the test film shall have a 
maximum lengthwise shrinkage of 0.50 per- 
cent when tested as follows: At least 20 strips 
of film approximately 31 inches in length 
shall be cut for measurement of shrinkage. 
After normal development and drying (not 
over +80F [+26.7C]), the strips shall be 
placed at least V* inch apart in racks and 
kept for seven days in an oven maintained at 
-f 1 20 F (+ 49 C) and a relative humidity of 
20 percent. The strips shall then be removed, 
reconditioned thoroughly to 50 percent rela- 
tive humidity at +70F (+21.1 C), and the 
shrinkage measured by an adaptation of the 
pin-gage method outlined in Research Paper 
RP-1051 of the National Bureau of Stand- 
ards. The percent shrinkage shall then be cal- 
culated on the basis of deviation from the 
nominal dimension for the length of 100 con- 
secutive perforation intervals given in Ameri- 
can Standard Z22. 12-1 941, or latest revision 
thereof. 

2.3 Standard Length of Film. The 

standard length of the flutter test film shall 
be 380 feet. 

2.4 Leader and Trailer. Each test film 
shall be furnished with a suitable leader, title, 
and trailer. 



NOTE: A test film in accordance with this stand- 
ard is available from the Society of Motion Picture 
Engineers. 



296 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard Specification for 

Multi-Frequency Test Film for Field Testing 

1 6-Millimeter Sound Motion Picture Projection Equipment 



R^f. U. S. fat. Off. 

Z22.44-1946 



1. Scope and Purpose 

1.1 This standard describes a multi-fre- 
quency sound test film used for testing and 
adjusting the sound systems of 16-mm sound 
motion picture projection equipment. The test 
frequencies on this film are adequate for nor- 
mal field and general laboratory use. 

2. Test Film 

2.1 Frequencies. The test film shall con- 
tain the following series of frequencies, each 
preceded by spoken announcement recorded 
at approximately 10 db below full modula- 
tion: 

Frequency Tone Footage 

Cycles Feet 



400 

50 

100 

200 

300 

500 

1000 

2000 

3000 

4000 

5000 

6000 

7000 

400 



12 
6 
6 
6 
6 
6 
6 
6 
6 
6 
6 
6 
6 

12 



2.1.1 Frequency Tolerance. The fre- 
quency tolerance of the recorded signals 
shall be 2 percent of the nominal fre- 
quency of each portion of the test track. 

2.2 Recording. The test film shall be an 
originally recorded, splice-free, direct play- 
back, positive variable-area sound track, re- 
corded so that the modulated light is substan- 
tially constant when the film is reproduced 
with a scanning beam of negligible width. 
Modulation of the recording shall be 95 =fc 5 
percent at 7000 cycles. The level within any 
one frequency of each reel shall be constant 
to within 0.5 db. The recording shall be 
accomplished on a recorder so constructed as 
to keep the flutter content of the film to the 
absolute minimum consistent with the state of 
the art. The distortion of the recorded wave, 



up to a frequency of 3000 cycles, shall not 
exceed 5 percent. 

2.3 Film Stock. The film stock used for the 
test film shall be cut and perforated in ac- 
cordance with the American Standard 16- 
Mm Sound Motion Picture Film; Cutting and 
Perforating Negative and Positive Raw Stock, 
Z22. 12-1 941, or latest revision thereof. 

2.3.1 Resistance to Shrinkage. The 
film stock used for the test film shall have a 
maximum lengthwise shrinkage of 0.50 per- 
cent when tested as follows: At least 20 
strips of film approximately 31 inches in 
length shall be cut for measurement of shrink- 
age. After normal development and drying 
(not over +80F [+26.7C]), the strips shall 
be placed at least V* inch apart in racks and 
kept for 7 days in an oven maintained at 
+ 1 20 F (-f 49 C) and a relative humidity of 
20 percent. The strips shall then be removed, 
reconditioned thoroughly to 50 percent rela- 
tive humidity at +70F (+21.1 C), and the 
shrinkage measured by an adaptation of the 
pin-gage method outlined in Research Paper 
RP-1051 of the National Bureau of Stand- 
ards. The percent shrinkage shall then be cal- 
culated on the basis of deviation from the 
nominal dimension for the length of 100 con- 
secutive perforation intervals given in Ameri- 
can Standard Z22.12-1941, or the latest re- 
vision thereof, referred to in 2.3 above. 

2.4 Film Identification. Each test film 
shall be provided with a suitable leader, title, 
and trailer, and shall be accompanied by a 
calibration of the level of the frequency re- 
cordings. 

2.4.1 Calibration. The calibration shall 
be in terms of light modulation at the photo- 
cell with a scanning beam of negligible width, 
and shall be correct to within =b } /4 db up to 
and including 3000 cycles, and within it Va 
db above 3000 cycles up to and including 
7000 cycles. The correction for each fre- 
quency shall be so stated that it will give the 
true level when the correction is added alge- 
braically to the output level measured using 
the film. 

NOTE: A test film in accordance with this stand- 
ard is available from the Society of Motion Picture 
Engineers. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



297 



American Standard Specifications for 

400-Cycle Signal Level Test Film 

for 16-Millimeter Sound Motion Picture Projection Equipment 



K-A'. V . S. Pal. Off. 

Z22.45-1946 



1. Scope and Purpose 

1.1 This specification describes a 400-cycle 
signal level test film for use in testing 16-mm 
sound motion picture projection equipment. 

2. Test Film 

2.1 Recording. The test film shall have an 
originally recorded direct playback positive 
variable-area sound track recorded at an 
amplitude of 0.048 =fc 0.0015 inch. The fre- 
quency of the recording shall be within 8 
cycles of nominal frequency. 

The density of the dark portion of the 
sound track shall be between 1.7 and 2.0. 
The density throughout the length of the film 
shall be as uniform as is consistent with the 
state of the art. 

The combined base and fog density, meas- 
ured as visual diffuse density, shall be 0.05 
0.01 when measured on an integrating 
sphere densitometer or a polarization den- 
si to meter. 

The wave form distortion of the recording 
shall not exceed 5 percent. 

2.2 Film Stock. The film stock used for the 
test film shall be cut and perforated in ac- 
cordance with the Amer^an Standard 16- 
Mm Sound Motion Picture Film; Cutting and 
Perforating for Negative and Positive Raw 
Stock, Z22.12-1941, or latest revision 
thereof. 

2.2.1 Res/stance to Shrinkage. The 
film stock used for the test film shall have a 
maximum lengthwise shrinkage of 0.50 per- 
cent when tested as follows: At least 20 strips 
of film approximately 31 inches in length 
shall be cut for measurement of shrinkage. 



After normal development and drying (not 
over +80F [+26.7C]), the strips shall be 
placed at least !4 inch apart in racks and kept 
for seven days in an oven maintained at 
-f 1 20 F ( -j- 49 C) and a relative humidity of 
20 percent. The strips shall then be removed, 
reconditioned thoroughly to 50 percent rela- 
tive humidity at +70F (+21.1 C), and the 
shrinkage measured by an adaptation of the 
pin-gage method outlined in Research Paper 
RP-1051 of the National Bureau of Stand- 
ards. The percent shrinkage shall then be cal- 
culated on the basis of deviation from the 
nominal dimension for the length of 100 con- 
secutive perforation intervals given in Ameri- 
can Standard Z22.12-1941, or latest revision 
thereof. 

2.3 Standard Length. The standard 
length of the 400-cycle signal level test film 
shall be 100 feet. 

2.4 Measurement Requirements. Each 
film shall be measured for amplitude of the 
modulation, for image density, and for com- 
bined base and fog density. The measure- 
ments shall be made at a point approxi- 
mately mid-length of the film and at points 
between 5 and 10 feet from each end. The 
results of the measurements shall be stated 
on a card furnished with each test film. 

2.5 Leader and Trailer. Each test film 
shall be provided with a suitable leader, title, 
and trailer. 



NOTE: A test film in accordance with this stand- 
ard is available from the Society of Motion Picture 
Engineers. 



298 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard 

16-Millimeter Positive Aperture Dimensions and 

Image Size for Positive Prints Made from 

35-Millimeter Negatives 



Reg. V. S. rat. Off. 

Z22.46-1946 













x~~ ' 






~~~~\ 

A > 




.^PRINTER 
-^APERTURE 








B > 


c 
v__ 


: t 


> 

H 

OF PRINT 

OF REDUC 
35 MM CAMf 


" REDUCED IMAGED 
OF 35 MM \. 
CAMERA APERTURE > 

1M 
^R APERTURE 
b IMAGE OF 

:RA APERTURE 










^ ' 



* REDUCTION RATIO 2.15 TO 1.00 



Millimeters 



Inch 



10.465 0.051 

10.211 0.051 

7.772 0.051 

7.41 7 0.038 



0.412 0.002 
0.402 0.002 
0.306 0.002 
0.292 0.0015 



*The tolerance of the specified reductfon ratio is expressed in the tolerance of the 
reduced image of the area covered by the 35-mm camera aperture. The dimensions of the 
35-mm camera aperture being reduced and of the 35-mm aperture of the reduction 
printer are 0.631 X 0.868 and 0.662 X 0.890 inch, respectively. 

NOTE 1: The reduced 16-mm image of the 35-mm camera aperture shall be centered in 
the 16-mm aperture of the printer. With the specified reduction ratio of 2.15, the equality 
of width of the two black lines (approximately 0.005 inch) produced on the print at the 
two sides of the aperture is an indication of proper centering. 

NOTE 2: Aperture corners may be rounded with a radius of 0.020 inch or less. 



April, 1946 



AMERICAN STANDARDS ON'MOTION PICTURES 



299 



American Standard 

Negative Aperture Dimensions and Image Size 

for 16-Millimeter Duplicate Negatives Made 

from 35-Millimeter Positive Prints 



Rep. V. S. I'at. Off. 

Z2 2. 47 -1946 
















"\ 


.^PRINTER 
^APERTURE 








( 


( 
OF 

OF PRINT 

OFREDUCI 
35-MM CAW 


REDUCED IMAGE -\ 
OF 35 MM \^ 
CAMERA APERTURE 

^ILM 
^R APERTURE 

TJ IMAGE OF 
ERA APERTURE 

- -/ 






t_T 



* REDUCTION RATIO 2.15 TO 1.00 



Millimeters 



Inch 



10.262 db 0.051 

10.211 0.051 

7.417 0.038 



0.404 0.002 
0.402 0.002 
0.292 0.0015 



*The tolerance of the specified reduction ratio is expressed in the tolerance of the 
reduced image of the area covered by the 35-mm camera aperture. The dimensions of the 
35-mm camera aperture being reduced and of the 35-mm aperture of the reduction printer 
are 0.631 X 0.868 and 0.662 X 0.890 inch, respectively. 

NOTE 1: The reduced 16-mm image of the 35-mm camera aperture shall be centered in 
the 16-mm aperture of the printer. 

NOTE 2: Aperture corners may be rounded with a radius of 0.020 inch or less. 



300 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard 

Printer Aperture Dimensions for Contact Printing 
16-Millimeter Positive Prints from 
16-Millimeter Negatives 


Rep. V. S. I'at. Off. 

Z22.48-1946 
































/-PRINTER 
^APERTURE 


1 


J 

< 
OF F 

OF PRIN1 


r ILM 
ER APERTURE 










I J 










Millimeters Inch 






A 10.465 0.051 0.41 2 0.002 
*B 7.772 0.051 0.306 0.002 



*This dimension is only applicable when using this aperture for contact printing by the 
step process. 

NOTE: Aperture corners may be rounded with a radius of 0.020 inch or less. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



301 



American Standard 

Printer Aperture Dimensions for 

Contact Printing 16-Millimeter Reversal 

and Color Reversal Duplicate Prints 



Reg. U. S. Pat. Off. 

Z22.49-1946 



OF FILM 
OF PRINTER APERTURE 



.-PRINTER 
'APERTURE 



Millimeters 



Inch 



10.262 0.051 
7.417 0.038 



0.404 0.002 
0.292 0.0015 



*This dimension is only applicable when using this aperture for contact printing by the 
step process. 

NOTE: Aperture corners may be rounded with a radius of 0.020 inch or less. 



302 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



American Standard 

Reel Spindles 

for 1 6-Millimeter Motion Picture Projectors 



Reg. U. S. Pat. Off. 

Z22.50-1946 



1. Round Section 

1.1 The round section of 16-mm motion pic- 
ture projector reel spindles shall have a fin- 
ished diameter of 0.312 0.003 inch (7.925 
0.076 mm). 

2. Square Section 

2.1 The square section of 16-mm motion 
picture projector reel spindles, including fin- 
ish, shall be 0.312 0.003 inch (7.925 
0.076 mm) across the flats. Measurements 
across the flats shall be made in mutually 
perpendicular directions. 

3. Cumulative Effect of 
Eccentricity 

3.1 The cumulative effect of eccentricity of 
the round and square sections of the spindles, 
looseness and misalignment of the bearing, 
or other mechanical imperfections shall not 
cause the flange of a tight-fitting reel to de- 
part from the ideal plane by more than 40 
minutes of arc. 



3.2 A suitable gage for determining the 
cumulative effect of eccentricity Consists of a 
hub, with coaxial square and round holes 
whose respective sides and diameter are 
equal in length, and a flange of suitable 
stiffness whose diameter is equal to that of 
an 800-foot reel flange, 10.5 inches (266.7 
mm). The flange should be permanently 
joined to the hub so that its face is perpen- 
dicular to the axis of the hub with not more 
than 0.003 inch (0.076 mm) runout. The hub 
shall be provided with a thumbscrew for 
clamping the hub to the reel spindle so that 
one side of the round and square holes shall 
come in contact with the corresponding 
round and square sections of the reel spindle. 

4. Reel Position on Spindles 

4.1 The design of spindles shall be such that 
reels are kept under constant lateral pres- 
sure against a shoulder on the spindle. The 
part forming this shoulder need not be integ- 
ral with the spindle. However, in such event, 
it shall be securely fastened to the spindle so 
so that the two parts rotate together. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



303 



American Standard Method of Making 

Intermodulation Tests on Variable-Density 
16-Millimeter Sound Motion Picture Prints 



K-g. 11. S. I'at: Off. 

Z22.51-1946 



1. Scope and Purpose 

1.1 This standard describes the intermodu- 
lation method of measuring the sound distor- 
tion introduced during the processing of 16- 
mm variable-density sound motion picture 
release prints. Through measurements of dis- 
tortion at various print densities, it is possible 
to choose a print density which will give 
sound prints having minimum distortion and 
hence optimum quality for the particular 
method of processing employed. 

1.2 In general, the intermodulation method 
of measuring distortion differs from the har- 
monic method in that the former employs a 
low frequency and a high frequency at one 
fourth the amplitude of the low frequency, 
the combination being simultaneously re- 
corded on the sound negative. Any distortion 
in the over-all process causes a change in 
high-frequency amplitude in portions of the 
low-frequency cycle. The ratio of the average 
variation in amplitude of the higher fre- 
quency in the reproduced wave to its origi- 
nal amplitude is called the intermodulation. 
Intermodulation test results are not directly 
proportional to harmonic measurements but 
in most cases an intermodulation figure of 
10 percent corresponds to a harmonic read- 
ing of about 2'/2 percent. 

2. Test Method 

2.1 The test track, as described in 2.3, is 
recorded and developed under standard 
conditions for the process being checked, 
and a series of prints made at suitable printer 
lights to give a range of print densities for 
unmodulated, unbiased tracks above and be- 
low the expected optimum density. In most 
cases this will be in the density range from 
0.4 to 0.7. There should be sufficient un- 
spliced film ahead of the test track to permit 
stabilization of printer speed. The distortion 
content of the test-track print is then meas- 
ured using equipment as shown in Fig. 1. In 
making the measurements, the test-track print 
is threaded through the sound head of the 



film reproducing device in the proper manner 
according to the position of the sound track 
(standard or nonstandard position). The dis- 
tortion of each section of the print is then 
measured according to the operating instruc- 
tions for the intermodulation meter user 1 




REPRODUCER AMPLIFIER 

PHOTOCELL 



MATCHING 
NETWORK 

(IF NEEDED) 



INTER- 
MODULATION 
METER 



Fig. 1 

Arrangement of Intermodulation Test 

Apparatus to Determine Distortion 

on Test Track 

2.2 Since the method here described meas- 
ures the over-all distortion for a process in- 
volving numerous variables, each of which 
may affect the total distortion, it is necessary 
that all such variables (except the print dens- 
ity which is purposely varied to find the 
optimum) in the recording and processing of 
the test track be maintained at the same val- 
ues as they are normally in the process to be 
checked. These variables include film stock 
(both sound negative and print), recorder- 
lamp current, negative gamma, positive 
gamma, color temperature of printer light, 
and type of printer (contact or optical). 

2.3 Test Track. The test track shall consist 
of 2 sections, recorded in sequence at the 
same lamp-current setting. 

2.3.1 Section 7. Section 1 shall be a 
combination of 60 cycles and 1 000 cycles hav- 
ing a peak amplitude 2.0 0.5 db below full 
modulation, in which the 1000-cycle level 
shall be 12 dr 1 db below the 60-cycle level. 
The 60- and 1000-cycle waves shall be not 
more than db 3 percent from the respective 
nominal frequencies and neither shall con- 
tain more than 5 percent harmonic distor- 
tion. This portion of the track should play 
for about 10 seconds. 

2.3.2 Section 2. Section 2 shall be at 
least 1 foot of unbiased, unmodulated track 
for density measurement. 



304 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 




MATCHING 

NETWORK 

(IF NEEDED) 



INTERMODULATION 
METER 



60- CYCLE 
SIGNAL GENERATOR 



Fig. 2 

Arrangement of Intermodulation Test Apparatus 
to Determine Distortion in Amplifier 




1000 -CYCLE 

BANDPASS 

FILTER 



Fig. 3 
Block Diagram of Intermodulation Meter 



3. Test Equipment 

3.1 Reproducing System. Care should 
be taken that the reproducer photocell and 
its associated coupling circuit to the ampli- 
fier input tube do not introduce a significant 
amount of intermodulation. The complete re- 
producing system should be checked for indi- 
cations of distortion by use of a suitable test 
film, when available. 

3.2 Amplifier. The amplifier of the film 
reproducer in which the test strips are run 
shall not produce more than 2 percent in- 
termodulation when the intermodulation fre- 
quencies are introduced directly into the 
input, as shown in Figure 2, at approximately 
the same level as that used when measuring 
intermodulation on a film. 

3.3 Intermodulation Meter. A block 
diagram of a suitable intermodulation 
meter* is shown in Figure 3. The attenuation 

The model RA-1107 intermodulation meter made 
by the Western Electric Company has been found 
suitable for these measurements. 



of the band-pass filter shall not vary more 
than 0.5 db between 880 and 1 1 20 cycles. 
The attenuation at 60 cycles shall be not less 
than 50 db and the attenuation at 500 cycles 
and at 2000 cycles not less than 25 db. The 
rectifier shall be such that its direct-current 
output is proportional to the peak amplitude 
of the wave passed by the band-pass filter. 
The low-pass filter following the rectifier shall 
have an attenuation that shall vary not more 
than db 0.5 db between 60 and 120 cycles 
and shall be at least 40 db at frequencies be- 
tween 1000 and 6000 cycles. The output 
indicating instrument shall be calibrated to 
read the average amplitude of the inter- 
modulation components in percentage of the 
average amplitude of the wave passed by 
the band-pass filter. 



References 

Frayne, J. G. and Scoville, R. R.: Analysis and meas- 
urement of distortion in variable density recording. 
Journal of the Society of Motion Picture Engineers, 
XXXII, No. 6 (June 1939), p 684. 



April, 1046 



AMERICAN STANDARDS ON MOTION PICTURES 



305 



American Standard Method of Making 

Cross-Modulation Tests on Variable-Area 
16-Millimeter Sound Motion Picture Prints 



Hi-?, t/./i. I'm. Uff. 

Z22.52-1946 



1. Scope and Purpose 

1.1 This standard describes the cross-modu- 
lation method of measuring high-frequency 
distortion introduced during the processing 
of 16-mm variable-area sound motion picture 
release prints. Through measurements of dis- 
tortion at various print densities, it is possible 
to choose a print density which will give 
sound prints having minimum distortion and 
optimum quality under the particular method 
of processing employed. 

1.2 Cross-modulation tests are used exten- 
sively in establishing and maintaining photo- 
graphic control of variable-area sound tracks. 
In consideration of wave form, output level, 
and noise reduction, it is necessary for vari- 
able-area prints to have high density con- 
trast. At satisfactorily high track densities an 
appreciable amount of image spread occurs, 
producing partial rectification of high fre- 
quencies. To compensate for this, an equal 
and opposite amount of image spread is in- 
troduced into the negative. Therefore, to 
establish the correct negative and print dens- 
ity combination, amplitude-modulated high 
frequencies are recorded and printed over a 
suitable density range. By measuring the rec- 
tified component from the prints, the correct 
density 'combinations are indicated. 



2. Test Method 

2.1 The test track, described in 2.3, is re- 
corded and developed under standard con- 
ditions for the process being checked. A 
series of prints is then made at suitable 
printer lights to give a range of print densi- 
ties for unmodulated, unbiased tracks above 
and below the expected optimum density. In 
most cases this will be in the density range 
from 0.8 to 1.7. There should be sufficient 
unspliced film ahead of the test track to per- 
mit stabilization of printer speed. The distor- 
tion content of the test-track print is then 



measured using equipment as shown in 
Fig. 1. 




400- CYCLE OUTPUT 

BAND-PASS METER 

FILTER 



Fig. 1 

Basic Arrangement of Cross-Modulation 

Test Equipment to Determine Distortion 

on Test Track 

In making the measurements, the test-track 
print is threaded through the sound head of 
the film reproducing device in the proper 
manner according to the position of the 
sound track (standard or non-standard posi- 
tion). The distortion of each section of the 
pHnt is then measured according to the oper- 
ating instructions for the cross-modulation 
meter used if one is available. 

2.1.1 If measurements are made using 
individual pieces of equipment, as shown in 
Fig. 1, the gain is adjusted to db output 
reading on the 400-cycle section of the test 
track, so that the output meter will properly 
read the db output from the 4000-cycle sec- 
tion of test track which is modulated by the 
400-cycle signal. The difference in output for 
the 2 sections of the cross-modulation test 
track is the cross-modulation for the process 
being checked. 

2.2 Since the method here described meas- 
ures the over-all distortion for a process in- 
volving numerous variables, each of which 
may affect the total distortion, it is necessary 
that all such variables (except the print dens- 
ity which is purposely varied to find the opti- 
mum) in the recording and processing of the 
test track be maintained at the same values 
as they are normally in the process to be 
checked. These variables include film stock 
(both sound negative and print), recorder- 
lamp current, negative gamma, color tem- 
perature of printer light, and type of printer 
(contact or optical). 



306 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



2.3 Test Track. The test track consists of 3 
sections, recorded in sequence at the same 
lamp-current setting: 

2.3.1 Section I. 400 10 cycles con- 
stant amplitude at 2 it Vi db below full 
modulation. 

2.3.2 Section 2. 4000 200 cycles 
amplitude modulated 2 d= Vi db below full 
modulation at 400 d= 10 cycles. The peak 
amplitude of this modulated wave shall be 2 
!/2 db below that required for full modula- 
tion of the sound track. 

2.3.3 Section 3. Unmodulated, un- 
biased track, or track fully modulated at 
approximately 30 cycles, or less, for density 
measurement. 

2.3.4 Sections 1 and 2 should have a run- 
ning time of 10 seconds and section 3 at 
least 1 second. Neither the 400- or 4000- 
cycle waves shall contain more than 5 per- 
cent harmonic distortion. The modulated 
wave of section 2 must not contain a 400- 
cycle component greater than 50 db below 
the average 4000-cycle amplitude of the 
modulated wave. The recorder should be 
correctly adjusted for slit azimuth, track lo- 
cation, and focus. 

NOTE: This test track should also be used in the 
case of 35-mm negatives prepared for a study of 
optical reduction printing, instead of the tracks com- 
monly used in checking regular 35-mm print produc- 
tion. 

3. Test Equipment 

3.1 Reproducing Equipment. The re- 
producing equipment shall consist of a 16- 
n'm sound film reproducer, amplifier, 400- 
cycle band-pass filter and volume indicator. 



3.2 Amplifying System. The amplifying 
system should have approximately 110 db 
gain and should be adjustable in 1 db steps 
over a range of 40 db. When the modulated 
wave described in 2.3.2 is applied to the 
input of the amplifier at sufficient amplitude 
to produce normal output, the 400-cycle com- 
ponent of the amplifier output shall be at 
least 50 db below the average 4000-cycle 
amplitude of the same modulated wave. The 
gain of the amplifying system shall be cali- 
brated at 400 and 4000 cycles. The relative 
gain at these frequencies shall vary not more 
than 1 db for any setting of the gain ad- 
justment. The attenuation of the 400-cycle 
band-pass filter shall vary not more than 
=fc 0.5 db between 375 and 425 cycles and 
shall be at least 50 db at 4000 cycles. 

3.3 Distortion. Care should be taken that 
the reproducer photocell and its associated 
coupling circuit to the amplifier input tube 
do not introduce a significant amount of 
cross-modulation. The complete reproducing 
system should be checked for indications of 
distortion by use of a suitable test film when 
available. 

References 

Baker, J. O. and Robinson, D. H.: Modulated high- 
frequency recording as a means of determining con- 
ditions for optimal processing. Journal of the Society 
of Motion Picture Engineers, XXX, No. 1 (Jan 
1938),p3. 

Trie Quality Control of Variable Area Sound Tracts. 
RCA Victor Division, Radio Corporation of America, 
Camden, N. J. (1938) 

Tfie Modulated Carrier Oscillator, RCA Victor Divi- 
sion, Radio Corporation of America, Camden, N. J. 
(1939) 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



307 



American Standard 

Method of Determining Resolving Power 

of 1 6-Millimeter Motion Picture Projector Lenses 



R'K. V. S. fat. Off. 

Z22.53-1946 



1. Scope and Purpose 

1.1 This standard describes a method of 
determining the resolving power of projec- 
tion lenses used in 16-mm motion picture 
projectors. The resolving power shall be 
measured in lines per millimeter. 

2. Test Method 

2.1 The lens to be tested shall be mounted 
in a special test projector. A glass plate test 
object, carrying patterns of lines, shall be 
then projected upon a white matte grainless 
screen located at such a distance from the 
projector that the projected image of the 
border of the test object measures 30 x 40 
inches. The resolving power of the lens is the 
largest number of lines per millimeter in the 
test object pattern that an observer standing 
close to the screen sees definitely resolved in 
both the radial and tangential directions. 
Lines* shall not be regarded as definitely re- 
solved unless the number of lines in the image 
is the same as the number of lines in the test 
object. 

2.1.1 The patterns of lines shall consist 
of parallel black lines 2.5/X mm long and 
0.5/X mm wide with a clear space 0.5/X mm 
wide between the parallel lines, where X 
equals the number of lines per millimeter. 

2.2 Care shall be taken to insure that the 
screen is perpendicular to the projection axis 
and that the lens is so focused that the image 
of the center of the test plate is as sharp as 
possible. 

3. Test Projector 

3.1 The projector design shall be such that 
the glass plate test object is held in proper re- 
lation to the lens axis. It shall not heat the test 
plate to a temperature which may cause the 



plate to be fractured or otherwise damaged. 
The cone of light supplied by the projector 
shall completely fill the projection lens, 

4. Test Object 

4.1 The glass photographic plate used for 
making the test object and the lens used* in 
making the reduction of the master test chart 
shall have sufficiently high resolving power 
to insure clear definition of all lines in the 
patterns on the test object. 

4.2 The photographic reduction of the mas- 
ter test chart shall be such that the test object 
border has a height of 7.21 mm (0.284 inch) 
and a width of 9.65 mm (0.380 inch) with a 
radius of 0.5 mm (0.02 inch) in the corners, 
and such that the sets of lines in the reduced 
image are spaced 20, 30, 40, 50, 60, 80, 
and 90 lines per millimeter. 



50 111= 111= 
40111= i=80 
30111= -90 




Fig. 1 
Resolution Test Patterns 



4.3 The patterns on the test object shall be 
in accordance with Fig. 1 . 



AMERICAN STANDARDS ON MOTION PICTURES 



Vol 46, No. 4 



"]o = 



60 - 550 
80 '- =11140 
90 -=11130 



PROJECT TO 40 INCHES ON 
WHITE MATTE GRAINLESS -SCREEN 



60 1 5..I50 
60" EIII40 
90 =11130 



50".= -i60 50"'= "< 

40MIS "BO 40IIIS -1 

30111= 90 

111= 



111 = 



60 i- 5... 50 
60'- =11140 
90 =11130 



50.5 -60 
40 ins -.80 
30111=- 90 



Fig. 2. Resolving Power Test Object 



NOTE: The triangular edge patterns are to facilitate alignment of test plates in the projector. 



4.4 The position of the test patterns on the 
test object shall be in accordance with 
Fig. 2. 

4.5 Identification of the positions of the test 
patterns on the test object shall be in accor- 
dance with Fig. 3. 



NOTE: Glass test plates in accordance with this 
standard are available from the Society of Motion 
Picture Engineers. 



ft) 



Fig. 3. Identification of Test Patterns 
in Frame Area 

NOTE: When using a 2-inch focal length lens, B 
corresponds to 2 degrees from the axis, C corre- 
sponds to 4 degrees from the axis, D corresponds to 
5 degrees from the axis, E corresponds to 6 degrees 
from the axis, and F corresponds to 3 degrees from 
the axis. 



April, 1946 



AMERICAN STANDARDS ON MOTION PICTURES 



309 



American Standard Methdd of Determining 

Freedom from Travel Ghost 

in 1 6-Millimeter Sound Motion Picture Projectors 



ASA 



Kr~. V S. Hal Off 

Z22.54-1946 



1. Scope and Purpose 

1.1 This standard describes a method of 
determining freedom from travel ghost in 
16-mm sound motion picture projectors. 

2. Definition 

2.1 Travel ghost is a blurring effect seen on 
the screen and evidenced by vertical tails or 
light streaks added to the projected images 
of the transparent areas on the test film. It is 
caused by the projector shutter being out of 
synchronism with the intermittent mechanism, 
either by faulty adjustment or faulty design. 




Fig. 1 Typical Test Pattern 



3. Test Film 

3.1 The test film used for determining free- 
dom from travel ghost shall carry a pattern 



of small transparent areas upon a dark back- 
ground. There shall be at least six transpar- 
ent areas, three of which shall be located 
not farther than 1/32 inch from the top of 
the frame and three not farther than 1/32 
inch from the bottom of the frame. Four of 
the areas shall have their edges 1/32 inch 
from a side edge and either the top or bot- 
tom edge of the frame. The density of the 
transparent areas shall be less than 0.2 and 
the density of the dark background shall be 
greater than 2.2. 

3.1.1 Standard Length of film. The 
standard length of test film shall be 100 feet. 

3.1.2 Leader and Trailer. Each test 
film shall have a suitable leader, title, and 
trailer. 

4. Test Method 

4.1 A test film in accordance with 2.1 shall 
be projected at standard sound speed of 24 
frames per second ( rb 2 percent) upon a 
white matte screen, the projected image of 
the projector picture aperture being of such 
size that a screen brightness of 10-foot lam- 
berts is obtained with the projector shutter 
running, but with no film in the gate. The 
screen image of the test film shall be viewed 
from a distance equal to twice its width and 
the presence or absence of travel ghost 
noted. 



NOTE: A test film in accordance with this stand- 
ard is available from the Society of Motion Picture 
Engineers. 



REPORT OF THE MEMBERSHIP AND SUBSCRIPTION 
COMMITTEE* 



The membership of the Society of Motion Picture Engineers grew 
during 1945 from 1676 to 1966. This increase is consistent with the 
trend during the war years. While it is a healthy, normal growth, 
the Committee feels that in 1946 it could be materially accelerated. 
To a large extent the new members acquired during the last several 
years have consisted of men in the Armed Forces engaged in motion 
picture activities. However, it is obvious that there are still many 
competent engineers and scientists working in our industry who 
are not members and who undoubtedly would like to become mem- 
bers if they knew the requirements of admission. 

The recent addition of an Engineering Secretary and staff to the 
office of the Society will permit a broad expansion in its technical 
activities and services to the industry. 

You are, of course, familiar with the fact that membership in the 
Society includes a subscription to the JOURNAL which is issued 
monthly, notices of technical sessions, the opportunity of participat- 
ing on the various Committees of the Society, and the unusual 
chance to meet and become acquainted with a large number of men 
in all phases of the motion picture industry. 

The development and improvement of many devices and processes 
used in our industry, as a result of the war, will stimulate interest in 
the Society among new companies. One of the most important of 
these fields is television. 

The President and Officers of the Society have been successful 
recently in increasing the number and amount of financial contribu- 
tions of Sustaining members. This will result in much greater recog- 
nition of the Society by all phases of the industry. 

We feel that we are justified in calling* upon each member of the 
Society to assist us in accomplishing our objective by obtaining at 
least one new member in 1946. A suitable application blank for this 
purpose is included in each regular issue of the JOURNAL and we 

* Submitted Mar. 28, 1946. 
310 



MEMBERSHIP AND SUBSCRIPTION COMMITTEE REPORT 311 

earnestly ask you to cut it out and mail it promptly to a prospective 
member with whom you are acquainted. In doing this, be sure to 
offer to sponsor their application. If everyone will perform this ob- 
ligation, the report of the Committee a year hence will show the type 
of results that we all are convinced can be performed. 

JAMES FRANK, JR., 
Chairman 



OFFICERS AND GOVERNORS^OF THE SOCIETY 




HERBERT GRIFFIN 
Past-President 



JOHN A. MAURER 
Engineering Vice-President 



LOREN L. RYDER 

Executive Vice-President 




DONALD E. HYNDMAN 
President 




M. RICHARD BOYER 
Financial Vice-President 

312 



ARTHUR C. DOWNES 
Editorial Vice-President 



WILLIAM C. KUNZMANN 
Convention Vice-President 



OFFICERS AND GOVERNORS OF THE SOCIETY 



313 




CLYDE R. KEITH 
Secretary 



EARL I. SPONABLE 
Treasurer 



JOHN I. CRABTREE 
Governor 




FRANK E. CARLSON 
Governor 



ALAN W. COOK 
Governor 



CHARLES R. DAILY 
Governor 





JOHN G. FRAYNE 
Governor 



PAUL J. LARSEN 
Governor 



WESLEY C. MILLER 
Governor 



314 



OFFICERS AND GOVERNORS OF THE SOCIETY 




PETER MOLE 
Governor 



REEVE O. STROCK 
Governor 



WILLIAM A. MUELLER 
Governor 




FRANK E. CAHILL, JR. 

Chairman, 
Atlantic Coast Section 




HOLLIS W. MOYSE 

Chairman, 
Pacific Coast Section 



ATLANTIC COAST SECTION 

*FRANK E. CAHILL, JR., Chairman 

*CLYDE R. KEITH, Past- Chairman 

*JAMES FRANK, JR., Sec.-Treas. 

* ""HERBERT BARNETT *G. T. LORANCE *W. H. OFFENHAUSER, JR. 
**H. D. BRADBURY **J. A. NORLING *H. E. WHITE 

PACIFIC COAST SECTION 

*HOLLIS W. MOYSE, Chairman 
*CHARLES W. HANDLEY, Past-Chairman 
*S. P. SOLOW, Sec.-Treas. 

**G. M. BEST **P. E. BRIGANDI *H. W. REMERSCHEID 

*J. W. BOYLE *F. L. EICH **G. E. SAWYER 



* Term expires December 31, 1946; ** Term expires December 31, 1947. 



COMMITTEES OF THE SOCIETY 

(Correct to March 26, 1946) 



ADMISSIONS. To pass upon all applications for membership, applications for transfer and 
to review the Student and Associate membership list periodically for possible transfers to the 
Associate and Active grades, respectively. The duties of each committee are limited to applica- 
tions and transfers originating in the geographic area covered. 

(East Coast) 
A. S. DICKINSON, Chairman 

28 West 44th St. 
New York 18, N. Y. 

M. R. BOYER JAMES FRANK, JR. HARRY RUBIN 

F. E. CAHILL, JR. J. A. MAURER E. I. SPONABLE 

(West Coast) 
H. W. REMERSCHEID, Chairman 

716 N. LaBrea St. 
Hollywood, Calif. 

C. R. DAILY PETER MOLE 

EMERY HUSE H. W. MOYSE 

BOARD OF EDITORS. To pass upon the suitability of all material submitted for publica- 
tion, or for presentation at conventions, and publish the JOURNAL. 

A. C. DOWNES, Chairman 

Box 6087 
Cleveland I, Ohio 

J. I. CRABTREE A. M. GUNDELFINGER C. R. KEITH 

A. N. GOLDSMITH C. W. HANDLEY E. W. KELLOGG 

A. C. HARDY 

CINEMATOGRAPHY. To make recommendations and prepare specifications for the 
operation, maintenance, and servicing of motion picture cameras, accessory equipment, studio 
and outdoor set lighting arrangements, camera technique, and the varied uses of motion picture 
negative films for general photography. 

J. W. BOYLE, Chairman 

1207 N. Mansfield Ave. 
Hollywood, Calif. 

C. G. CLARKE * ARTHUR MILLER ARTHUR REEVES 
KARL FREUND JOSEPH RUTTENBERG 

COLOR. To make recommendations and prepare specifications for the operation, mainte- 
nance, and servicing of color motion picture processes, accessory equipment, studio lighting, 
selection of studio set colors, color cameras, color motion picture films, and general color photog- 
raphy. 

J. A. BALL, Chairman 

12720 Highwood St. 
Los Angeles 24, Calif. 



M. R. BOYER R. M. EVANS A. C. HARDY 

G. A. CHAMBERS J. L. FORREST W. C. MILLER 

L. E. CLARK J. G. FRAYNB L. L. RYDER 

R. O. DREW L.' T. GOLDSMITH J. G. WHITE 

A. M. GUNDELFINGER 



* Advisory Member. 

315 



316 COMMITTEES OF THE SOCIETY Vol 46, No. 4 

CONVENTION. To assist the Convention Vice-President in the responsibilities pertaining 
to arrangements and details of the Society's technical conventions. 

W. C. KUNZMANN, Chairman 
Box 6087 
Cleveland 1, Ohio 

*JULIUS HABER C. R. KEITH H. W. MOYSE 

H. F. HEIDEGGER R. H. McCuixouGH O. F. NEU 

EXCHANGE PRACTICE. To make recommendations and prepare specifications on the 
engineering or technical methods and equipment that contribute to efficiency in handling and 
storage of motion picture prints, so far as can be obtained by proper design, construction, and 
operation of film handling equipment, air-conditioning systems, and exchange office buildings. 

(Under Organization) 

FELLOW MEMBERSHIP. To consider qualifications of Active members as candidates for 
elevation to Fellow members, and to submit such nominations to the Board of Governors. 

HERBERT GRIFFIN, Chairman 

133 E. Santa Anita Ave. 
Burbank:, Calif. 

M. R. BOYER D. E. HYNDMAN H. W. MOYSE 

A. S. DICKINSON C. R. KEITH L. L. RYDER 

A. C. DOWNES W. C. KUNZMANN E. A. WILLIFORD 

J. A. MAURER 

FILM PROJECTION PRACTICE. To make recommendations and prepare specifications 
for the operation, maintenance and servicing of motion picture projection equipment, projection 
rooms, film storage facilities, stage arrangement, screen dimensions and placement, and main- 
tenance of loudspeakers to improve the quality of reproduced sound and the quality of the 
projected picture in the theater. 

G. T. LORANCE, Chairman 

63 Bedford Road 
Pleasantville, N. Y. 

HENRY ANDERSON L. W. DAVEE C. F. HORSTMAN 

*C. J. BACHMAN A. S. DICKINSON L. B. ISAAC 

T. C. BARROWS J. K. ELDERKIN E. R. MORIN 

H. D. BEHR JAMES FRANK, JR. M. D. O'BRIEN 

H. J. BENHAM R. R. FRENCH HARRY RUBIN 

M. F. BENNETT G. GAGLIARDI J. J. SEFING 

KARL BRENKERT E. R. GEIB R. O. WALKER 

F. E. CAHILL, JR. ADOLPH GOODMAN V. A. WELMAN 

D. W. COLLINS HERBERT GRIFFIN H. E. WHITE 

C. C. DASH SYLVAN HARRIS A. T. WILLIAMS 
J. J. HOPKINS 

HISTORICAL AND MUSEUM. To collect facts and assemble data relating to the historical 
development of the motion picture industry, to encourage pioneers to place their work on record 
in the form of papers for publication in the JOURNAL, and to place in suitable depositories equip- 
ment pertaining to the industry. 

J. E. ABBOTT, Chairman 

11 West 53d St. 
New York 19, N. Y. 

O. B. DEPUE RICHARD GRIFFITH TERRY RAM SAVE 

HONORARY MEMBERSHIP. To diligently search for candidates who through their 
basic inventions or outstanding accomplishments have contributed to the advancement of the 
motion picture industry and are thus worthy of becoming Honorary members of the Society. 

EMERY HUSB, Chairman 

6706 Santa Monica Blvd. 
Hollywood 38, Calif. 

NATHAN LEVINSON L. L. RYDER 

A. J. MILLER E. I. SPONABLB 



* Advisory Member. 



April, 1946 COMMITTEES OF THE SOCIETY 317 

JOURNAL AWARD. To recommend to the Board of Governors the author or authors of 
the most outstanding paper originally published in the JOURNAL during the preceding calendar 
year to receive the Society's Journal Award. 

F. E. CARLSON, Chairman 

Nela Park 
Cleveland 12, Ohio 

C. R. DAILY G. E. MATTHEWS 

P. J. LARSEN W. V. WOLFB 

LABORATORY PRACTICE. To make recommendations and prepare specifications for the 
operation, maintenance, and servicing of motion picture printers, processing machines, inspec- 
tion projectors, splicing machines, film cleaning and treating equipment, rewinding equipment, 
any type of film handling accessories, methods, and processes which offer increased efficiency 
and improvement in the photographic quality of the final print. 

H. E. WHITE, Temporary Chairman 

Room 813 

350 Madison Ave. 

New York 17, N. Y. 

J. R. ALBURGER F. L. EICH J. M. NICKOLAUS 

A. C. BLANEY G. H. GIBSON N. F. OAKLEY 

L. A. BONN EMERY HUSE W. H. OFFENHAUSER. JR. 

A. W. COOK T. M. INGMAN V. C. SHANER 

O. B. DEPUE C. L. LOOTENS J. H. SPRAY 

R. O. DREW A. J. MILLER J. F. VAN LEUVEN 

MEMBERSHIP AND SUBSCRIPTION. To solicit new members, obtain nonmember sub- 
scriptions for the JOURNAL, and to arouse general interest in the activities of the Society and its 
publications. 

(East Coast) 
JAMES FRANK, JR.. Chairman 

356 West 44th St. 
New York 18, N. Y. 

T. C. BARROWS L. T. GOLDSMITH G. E. MATTHEWS 

J. G. BRADLEY SYVAN HARRIS G. C. MISENER 

KARL BRENKERT L. B. ISAAC H. B. SANTEE 

G. A. CHAMBERS W. C. KUNZMANN E. O. WILSCHKB 

E. R. GEIB S. A. LUKES C. R. WOOD, SR. 

(West Coast) 
H. W. REMERSCHEID, Chairman 

716 N. LaBrea St. 
Hollywood, Calif. 

L. W. CHASE, JR. HERBERT GRIFFIN L. L. RYDER 

J. P. CORCORAN EMERY HUSE G. E. SAWYER 

C. R. DAILY K. F. MORGAN W. L. THAYER 

J. G. FRAYNE H. W. MOYSE W. V. WOLFE 

W. A. MUELLER 

NOMINATIONS. To recommend nominations to the Board of Governors for annual election 
of officers and governors. 

E. M. HONAN, Chairman 

6601 Remain e St. 
Hollywood 38, Calif. 

EJ. A. BERTRAM EMERY HUSE W. C. MILLER 

M. R. BOYER D. B. JOY PETER MOLE 

HERBERT GRIFFIN J. A. MAURER E. A. WILLIFORD 



318 COMMITTEES OF THE SOCIETY Vol 46, No. 4 

PAPERS. To solicit papers, and provide the program for semi-annual conventions, and make 
availablt to local sections for their meetings papers presented at national conventions. 

BARTON KREUZER, Chairman C. R. DAILY, Vice-Chairman 

RCA Victor Division 5451 Marathon St. 

Radio Corp. of America Hollywood 38, Calif. 

Camden, N. J. 

A. R. DAVIS L. T. GOLDSMITH H. W. MOYSE 

F. L. EICH C. R. KEITH V. C. SHANER 

JAMES FRANK, JR. E. W. KELLOGG S. P. SOLOW 

J. G. FRAYNE G. E. MATTHEWS W. V. WOLFE 

P. A. McGuiRB 

PRESERVATION OF FILM. To make recommendations and prepare specifications on 
methods of treating and storage of motion picture film for active, archival, and permanent 
record purposes, so far as can be prepared within both the economic and historical value of the 
films. 

J. G. BRADLEY, Chairman 

The Library of Congress 
Washington 25, D. C. 

J.E.ABBOTT" J.L.FORREST *C. A. LINDSTROM 

*H. T. COWLING *J. E. GIBSON G. S. MITCHELL 

J. I. CRABTREB *ORVILLE GOLDNER TERRY RAMSAYE 

A. S. DICKINSON C. L. GREGORY V. B. SEASE 

PROCESS PHOTOGRAPHY. To make recommendations and prepare specifications on 
motion picture optical printers, process projectors (background process), matte processes, 
special process lighting technique, special processing machines, miniature set requirements, 
special effects devices, and the like, that will lead to improvement in this phase of the production 
art. 

(Under Organization) 

PROGRESS. To prepare an annual report on progress in the motion picture industry. 

W. V. WOLFE, Chairman 

1016 N. Sycamore Ave. 
Hollywood 38, Calif. 

(Under Organization) 

PROGRESS MEDAL AWARD. To recommend to the Board of Governors a candidate who 
by his inventions, research, or development has contributed in a significant manner to the 
advancement of motion picture technology, and is deemed worthy of receiving the Progress 
Medal Award of the Society. 

E. A. WILLIFORD, Chairman 

230 Park Ave. 
New York 17, N. Y. 

M. R. BOYER NATHAN LEVINSON 

F. E. CARLSON G. F. RACKETT 

PUBLICITY. To assist the Convention Vice-President in the release of publicity material 
concerning the Society's semi-annual technical conventions. 

*HAROLD DESFOR, Chairman 
RCA Victor Division 
Radio Corp. of America 
Camden, N. J. 

LEONARD BIDWELL C. R. DAILY P. A. McGuiRB 

E. O. BLACKBURN BARTON KREUZER HARRY SHERMAN 

SCREEN BRIGHTNESS. To make recommendations, prepare specifications, and test 
methods for determining and standardizing the brightness of the motion picture screen image 
at various parts of the screen, and for special means or devices in the projection room adapted 
to the control or improvement of screen brightness. 

E. R. GEIB, Chairman 
National Carbon Company, Inc. 
Fostoria Works 
Fostoria, Ohio 



* Advisory Member. 



April, 1946 



COMMITTEES OF THE SOCIETY 



319 



HERBERT BARNETT 
F. E. CARLSON 
SYLVAN HARRIS 



W. F. LITTLE 
W. B. RAYTON 
C. M. TUTTLE 
C. R. UNDERBILL 



H. E. WHITE 
A. T. WILLIAMS 
R. J. ZAVESKY 



16-MM AND 8-MM MOTION PICTURES (Formerly Nontheatrical Equipment). To 
make recommendations and prepare specifications for 16-mm and 8-mm cameras, 16-mm 
sound recorders and sound recording practices, 16-mm and 8-mm printers and other film labo- 
ratory equipment and practices, 16-mm and 8-mm projectors, splicing machines, screen dimen- 
sions and placement, loudspeaker output and placement, preview or theater arrangements, 
test films, and the like, which will improve the quality of 16-mm and 8-mm motion picturest 



D. F. LYMAN, Chairman 

333 State St. 
Rochester 4, N. Y. 



R. C. HOLSLAG 

H. J. HOOD 

R. KlNGSLAKE 

L. R. MARTIN 
W. C. MILLER 
*A. H. NICOL 
V. J. NOLAN 

W. H. OFFENHAUSER, JR. 
M. W. PALMER 
A. G. PETRASEK 
S. READ, JR. 



E. W. D'ARCY 
WILLIAM BALCH 
W. C. BOWEN 

F. L. BRETHAUER 
*F. E. BROOKER 

F. E. CARLSON 

G. A. CHAMBERS 
S. L. CHERTOK 
JOHN CHRISTIE 
G. W. COLBURN 
R. O. DREW 

*J. WALTER EVANS 

SOUND. To make recommendations and prepare specifications for the operation, mainte- 
nance, and servicing of motion picture film, sound recorders, rerecorders, and reproducing 
equipment, methods of recording sound, sound film processing, and the like, to obtain means of 
standardizing procedures that will result in the production of better uniform quality sound in 
the theater. 

J. G. FRAYNE, Chairman C. R. KEITH, Vice-Chairman 

6601 Romaine St. 233 Broadway 

Hollywood 38, Calif. New York 7, N. Y. 



L. T. SACHTLEBEN 

A. SHAPIRO 

D. G. SMITH 

RAYMOND SPOTTISWOODE 
*J. B. STREIFFERT 

HARRY STRONG 
*A. L. TERLOUW 

LLOYD THOMPSON 

M. G. TOWNSLEY 

L. E. VARDEN 

J. E. VOLKMANN 

A. G. ZIMMERMAN 



D. J. BLOOMBERG 

B. B. BROWN 

F. E. CAHILL, JR. 

C. R. DAILY 
R. J. ENGLER 
L. D. GRIGNON 



L. B. ISAAC 
J. P. LIVADARY 

G. T. LORANCE 

W. C. MILLER 
W. A. MUELLER 



OTTO SANDVIK 
G. E. SAWYER 
S. P. SOLOW 
E. I. SPONABLE 
R. T. VANNIMAN 
J. E. VOLKMANN 



STANDARDS. To constantly survey all engineering phases of motion picture production, 
distribution, and exhibition, to make recommendations and prepare specifications that may 
become proposals for SMPE Recommended Practices and/or American Standards. This 
Committee should carefully follow the work of all other committees on engineering and may 
request any committee to investigate and prepare a report on the phase of motion picture 
engineering to which it is assigned. 



J. M. ANDREAS 
HERBERT BARNETT 
M. F. BENNETT 

E. A. BERTRAM 
M. R. BOYBR 

F. L. BRETHAUER 

F. E. CARLSON 
E. K. CARVER 

G. A. CHAMBERS 
J. S. CHANDLER 
A. W. COOK 

E. D. COOK 
L. W. DAVEB 



F. T. BOWDITCH, Chairman 

Box 6087 
Cleveland 1, Ohio 

L. T. GOLDSMITH 
IRL GOSHAW 
HERBERT GRIFFIN 
A. C. HARDY 
R. C. HOLSLAG 
J. K. HILLIARD 

C. S. PERKINSf 

D. B. JOY 
C. R. KEITH 

R. KlNGSLAKB 

P. J. LARSEN 
C. L. LOOTENS 



H. W. MOYSE 

W. H. OFFENHAUSER, JR. 

J. D. PHYFB 

G. F. RACKBTT 

W. B. RAYTON 

A. C. ROBERTSON 

L. T. SACHTLEBEN 

OTTO SANDVIK 

J. A. SCHBIK 

R. R. SCOVILLE 

J. H. SPRAY 

MONROE SWEET 



D. F. LYMAN 



* Advisory Member; t Alternate. 



LLOYD THOMPSON 

(See next page.) 



320 



COMMITTEES OF THE SOCIETY 



Vol 46, No. 4 



A. A. DURYEA 

A. F. EDOUART 

P. C. GOLDMARK 

A. N. GOLDSMITH 

STUDIO LIGHTING. To 



PIERRE MERTZ 
W. C. MILLER 
G. S. MITCHELL 



M. G. TOWNSLBY 
J. F. VAN LEUVEN 
D. R. 



fe. 



WHITE 
H. E. WHITE 

make recommendations and prepare specifications for the 
operation, maintenance, and servicing of all types of studio and outdoor auxiliary lighting 
equipment, tungsten light and carbon arc sources, lighting effect devices, diff users, special light 
screens, etc., to increase the general engineering knowledge of the art. 

C. W. HANDLEY, Chairman 

I960 West 84th St. 
Los Angeles 44, Calif. 

F. W. BOYLE F. C. COATES KARL FREUND 
[. J. CHANON R. E. FARNHAM W. W. LOZIER 

TECHNICAL NEWS. To survey the fields of production, distribution, and exhibition of 
motion pictures, and allied industries, to obtain technical news items for publication in the 
JOURNAL. 

A. C. BLANEY, Chairman 

1016 N. Sycamore St. 
Hollywood 38, Calif. 

C. W. HANDLEY 
EMERY HUSE 
H. R. LUBCKE 
K. F. MORGAN 

TELEVISION. To study the television art with special reference to the technical inter- 
relationships of the television and motion picture industries, and to make recommendations 
and prepare specifications for equipment, methods, and nomenclature designed to meet the 
special problems encountered at the junction of the two industries. 

D. R. WHITE, Chairman 

Redpath Laboratories 

E. I. du Pont de Nemours & Co. 
Parlin, N. J. 



J. W. BOYLB 

J. I. CRABTREE 

A. M. GUNDELFINGER 



H. W. REMERSCHEID 
EDWARD SCHMIDT 
WILLIAM THOMAS 



R. B. AUSTRIAN 
*G. L. BEERS 
A. W. COOK 
E. D. COOK 
C. E. DEAN 
BERNARD ERDE 
P. C. GOLDMARK 
A. N. GOLDSMITH 
T. T. GOLDSMITH 
HERBERT GRIFFIN 

*F. P. GOLDBACHf 



C. F. HORSTMAN 

L. B. ISAAC 
A. G. JENSEN 
P. J. LARSEN 
C. C. LARSON 
NATHAN LEVINSON 
F. E. CAHILL, jR 



J. P. LlVADARY 

H. B. LUBCKE 

J. R. POPPELEf 

PIERRE MERTZ 
W. C. MILLER 
*PAUL RAIBOURN 
OTTO SANDVIK 
G. E. SAWYER 
R. E. SHELBY 
E. I. SPONABLE 
H. E. WHITE 



TELEVISION PROJECTION PRACTICE. To make recommendations and prepare speci- 
fications for the construction, installation, operation, maintenance, and servicing of equipment 
for projecting television pictures in the motion picture theater, as well as projection room 
arrangements necessary for such equipment, and such picture-dimensional and screen-charac- 
teristic matters as may be involved in high-quality theater television presentation. 



P. J. LARSEN, Chairman 

1401 Sheridan St., N. W. 
Washington 11, D. C. 



F. E. CAHILL, JR., Vice-Chairman 

321 West 44th St. 
New York 18, N. Y. 



R. B. AUSTRIAN 

C. F. HORSTMAN f 
HERBERT BARNETT 

*F. P. GOLDBACHf 

M. C. BATSEL 

F. G. ALBINf 

*G. L. BEERS 



L. W. DAVEE 
JOHN EBERSON 
*H. B. FANCHER 

*P. M. GARRETTf 

JAMES FRANK, JR. 

G. T. LORANCEf 

E. P. GENOCK 



JAMES FRANK, JR., Secretary 

356 West 44th St. 
New York 18, N. Y. 

H. E. KALLMANN 
E. R. MORIN 

A. H. ROSENTHAL 

* ARTHUR LEVEY t 
HARRY RUBIN 

*PAUL RAIBOURN f 
R. E. SHELBY 



* Advisory Member; t Alternate. 



April, 1946 COMMITTEES OP THE SOCIETY 321 

F. G. ALBINf A. J. RICHARD f *E. D. GoODALEf . 

*THOMAS BILLS P. C. GOLDMARK E. I. SPONABLE 
F. T. BOWDITCH G. R. TiNGLEvf *H. E. BRAGG f 

PAUL RiEsf T. T. GOLDSMITH C. S. SZEGHO 
A. BROLLY *RUDOLPH FELDT| NICHOLAS GLYPTISJ 

*W. BROCKf L. B. ISAAC V. A. WELMAN 

F. E. CAHILL, JR. M. D. O'BRiENf H. E. WHITE 

M. F. BENNErrf A. G. JENSEN D. E. HYNDMANJ 

H. B. SANTEEf 

J. J. KOHLER 
PIERRE MERTZ 

TEST FILM QUALITY. To supervise, inspect, and approve all print quality control of 
sound and picture test films prepared by any committee on engineering before the prints are 
released by the Society for general practical use. 

F. S. BERMAN, Chairman 

111-14 76th Ave. 
Forest Hills, N. Y. 

C. F. HORSTMAN F. R. WILSON 

THEATER ENGINEERING, CONSTRUCTION AND OPERATION. To make recom- 
mendations and prepare specifications on engineering methods and equipment of motion picture 
theaters in relation to their contribution to the physical comfort and safety of patrons, so far 
as can be enhanced by correct theater design, construction, and operation of equipment. 

HBNRY ANDERSON, Chairman 

1501 Broadway 
New York 18. N. Y. 

HERBERT BARNETT *W. L. FLEISHER C. F. HORSTMAN 

H. J. BENHAM JAMES FRANK, JR. E. R. MORIN 

F. E. CARLSON T. T. GOLDSMITH BEN SCHLANGER 

*W. B. CUTTER ADOLPH GOODMAN J. J. SEFING 

J. J. HOPKINS 

SMPE REPRESENTATIVES TO OTHER ORGANIZATIONS 

American Documentation Institute J. E. ABBOTT 

American Standards Association: 

Sectional Committee on Standardization of Letter 
Symbols and Abbreviations for Science and 

Engineering, Z10 L. A JONES 

Sectional Committee on Motion Pictures, Z22 . .C. R. KEITH, Chm. 

A.N. GOLDSMITH, Hon. Chm 
F. T. BOWDITCH 

E. K. CARVER 

D F. LYMANf 

Sectional Committee on Acoustical Measurements 

and Terminology, Z24 J. E. VOLKMANN 

Sectional Committee on Photography, Z38 J. I. CRABTREE 

European Advisory Committee DONALD MCMASTER, Chm 

Inter-Society Color Council R. M. EVANS, Chm 

J. A. BALL 
RONALD BINGHAM 

F. T. BOWDITCH 
M. R. BOYER 

A. M. GUNDELFINGBR 

G. F. RACKBTT 

National Fire Protection Association A. S. DICKINSON 

Radio Technical Planning Board P. J. LARSBN 

E. I. SPONABLBf 

* Advisory Member; t Alternate. 



CONSTITUTION AND BY-LAWS* 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 

CONSTITUTION 
Article I 

Name 

The name of this association shall be SOCIETY OF MOTION PICTURE 
ENGINEERS. 

Article II 

Object 

Its objects shall be: Advancement in the theory and practice of motion pic- 
ture engineering and the allied arts and sciences, the standardization of the equip- 
ment, mechanisms, and practices employed therein, the maintenance of a high 
professional standing among its members, and the dissemination of scientific 
knowledge by publication. 

Article III 

Eligibility 

Any person of good character may be a member in any grade for which he is 
eligible. 

Article W 

Officers 

The officers of the Society shall be a President, a Past- President, an Executive 
Vice-President, an Engineering Vice-President, an Editorial Vice-President, a 
Financial Vice-President, a Convention Vice-President, a Secretary, and a 
Treasurer. 

The term of office of all elected officers shall be for a period of two years. Of 
the Engineering, Editorial, Financial, and Convention Vice-Presidents, and the 
Secretary, and the Treasurer, three shall be elected alternately each year, or until 
their successors are chosen. The President shall not be immediately eligible to 
succeed himself in office. Under such conditions as set forth in the By-Laws the 
office of Executive Vice-President may be vacated before the expiration of his 
term. 

Article V 

Board of Governors 

The Board of Governors shall consist of the President, the Past-President, the 
five Vice-Presidents, the Secretary, the Treasurer, the Section Chairmen and 

* Corrected to March 15, 1946. 
322 



CONSTITUTION AND BY-LAWS 323 

ten elected governors. Five of these governors shall be resident in the area operat- 
ing under Pacific and Mountain time, and five of the governors shall be resident 
in the area operating under Central and Eastern time. Two of the governors 
from the Pacific area and three of the governors from the Eastern area shall be 
elected in the odd-numbered years, and three of the governors in the Pacific area 
and two of the governors in the Eastern area shall be elected in the even-numbered 
years. The term of office of all elected governors shall be for a period of two 
years. 

Article VI 

Meetings 

There shall be an annual meeting, and such other meetings as stated in the 
By-Laws. 

Article VII 

Amendments 

This Constitution may be amended as follows: Amendments shall be approved 
by the Board of Governors, and shall be submitted for discussion at any regular 
members' meeting. The proposed amendment and complete discussion then shall 
be submitted to the entire Active, Fellow, and Honorary membership, together 
with letter ballot as soon as possible after the meeting. Two-thirds of the vote 
cast within sixty days after mailing shall be required to carry the amendment. 

BY-LAWS 
By-Law I 

Membership 

Sec. 1 . The membership of the Society shall consist of Honorary members. 
Fellows, Active members, Associate members, Student members, and Sustaining 
members. 

An Honorary member is one who has performed eminent services in the ad- 
vancement of motion picture engineering or in the allied arts. An Honorary 
member shall be entitled to vote and to hold any office in the Society. 

A Fellow is one who shall not be less than thirty years of age and who shall 
comply with the requirements of either (a) or (6) for Active members and, in 
addition, shall by his proficiency and contributions have attained to an out- 
standing rank among engineers or executives of the motion picture industry. 
A Fellow shall be entitled to vote and to hold any office in the Society. 

An Active member is one who shall be not less than 25 years of age, and shall 
be (a) a motion picture engineer by profession. He shall have been engaged in 
the practice of his profession for a period of at least three years, and shall have 
taken responsibility for the design, installation, or operation of systems or ap- 
paratus pertaining to the motion picture industry; (&) a person regularly em- 
ployed in motion picture or closely allied work, who by his inventions or pro- 
ficiency in motion picture science or as an executive of a motion picture enterprise 
of large scope, has attained to a recognized standing in the motion picture industry. 



324 CONSTITUTION AND BY-LAWS Vol 46, No. 4 

In case of such an executive, the applicant must be qualified to take full charge 
of the broader features of motion picture engineering involved in the work under 
his direction. 

An Active member is privileged to vote and to hold any office in the Society. 

An Associate member is one who shall be not less than 18 years of age, and shall 
be a person who is interested in or connected with the study of motion picture 
technical problems or the application of them. An Associate member is not privi- 
leged to vote, to hold office or to act as chairman of any committee, although he 
may serve upon any committee to which he may be appointed; and, when so 
appointed, shall be entitled to the full voting privileges of a committee member. 

A Student member is any person registered as a student, graduate or under- 
graduate, in a college, university, or educational institution, pursuing a course of 
studies in science or engineering that evidences interest in motion picture tech- 
nology. Membership in this grade shall not extend more than one year beyond 
the termination of the student status described above. A Student member shall 
have the same privileges as an Associate member of the Society. 

A Sustaining member is an individual, a firm, or corporation contributing sub- 
stantially to the financial support of the Society. 

Sec. 2. All applications for membership or transfer, except for Honorary or 
Fellow membership, shall be made on blank forms provided for the purpose, and 
shall give a complete record of the applicant's education and experience. Honor- 
ary and Fellow membership may not be applied for. 

Sec. 3. (a) Honorary membership may be granted upon recommendation 
of the Board of Governors when confirmed by a four-fifths majority vote of the 
Honorary members, Fellows, and Active members present at any regular meeting 
of the Society. An Honorary member shall be exempt from all dues. 

(&) Fellow membership may be granted upon recommendation of the Fellow 
Membership Award Committee, when confirmed by a three-fourths majority vote 
of the Board of Governors. Nominations for Fellow shall be made from the Active 
membership. 

(c) Applicants for Active membership shall give as references at least one mem- 
ber of Active or of higher grade in good standing. Applicants shall be elected 
to membership by the unanimous approval of the entire membership of the ap- 
propriate Admissions Committee. In the event of a single dissenting vote or 
failure of any member of the Admissions Committee to vote, this application shall 
be referred to the Board of Governors, in which case approval of at least three- 
fourths of the Board of Governors shall be required. 

(d) Applicants for Associate membership shall give as references one member 
of the Society in good standing, or two persons not members of the Society who 
are associated with the industry. Applicants shall be elected to membership 
by approval of a majority of the appropriate Admissions Committee. 

(e) Applicants for Student membership shall give as reference the head of the 
department of the institution he is attending, this faculty member not necessarily 
being a member of the Society. 

By-Law II 

Officers 

Sec. IAn officer or governor shall be an Honorary, a Fellow, or an Active 
member. 



April, 1946 CONSTITUTION AND B Y-LAWS 325 

Sec. 2. Vacancies in the Board of Governors shall be filled by the Board of 
Governors until the annual meeting of the Society. 

By-Law III 

Board of Governors 

Sec. 1. The Board of Governors shall transact the business of the Society 
between members' meetings, and shall meet at the call of the President, with the 
proviso that no meeting shall be called without at least seven (7) days' prior 
notice, stating the purpose of the meeting, to all members of the Board by letter or 
by telegram. 

Sec. 2. Nine members of the Board of Governors shall constitute a quorum 
at all meetings. 

Sec. 3. When voting by letter ballot, a majority affirmative vote of the total 
membership of the Board of Governors shall carry approval, except as otherwise 
provided. 

Sec. 4. The Board of Governors, when making nominations to fill vacancies 
in offices or on the Board, shall endeavor to nominate persons who in the aggregate 
are representative of the various branches or organizations of the motion picture 
industry to the end that there shall be no substantial predominance upon the 
Board, as the result of its own action, of representatives of any one or more 
branches or organizations of the industry. 

By-Law IV 

Committees 

Sec. 1. All committees, except as otherwise specified, shall be appointed by the 
President. 

Sec. 2. All committees shall be appointed to act for the term served by the 
officer who shall appoint the committees, unless their appointment is sooner ter- 
minated by the appointing officer. 

Sec. 3. Chan-men of the committees shall not be eligible to serve in such ca- 
pacity for more than two consecutive terms. 

Sec. 4. Standing committees of the Society shall be as follows to be appointed 
as designated: 

(a) Appointed by the President and confirmed by the Board of Governors 

Progress Medal Award Committee 
Journal Award Committee 
Honorary Membership Committee 
Fellow Membership Award Committee 
Admissions Committees 

(Atlantic Coast Section) 

(Pacific Coast Section) 
European Advisory Committee 

(b) Appointed by the Engineering Vice- President 

Sound Committee 
Standards Committee 



326 CONSTITUTION AND BY-LAWS Vol 46, No. 4 

Studio Lighting Committee 

Color Committee 

Theater Engineering Committee 

Exchange Practice Committee 

Nontheatrical Equipment Committee 

Television Committee 

Test Film Quality Committee 

Laboratory Practice Committee 

Cinematography Committee 

Process Photography Committee 

Preservation of Film Committee 

(c) Appointed by the Editorial Vice-President 

Board of Editors 
Papers Committee 
Progress Committee 
Historical Committee 
Museum Committee 

(d) Appointed by the Convention Vice-President 

Publicity Committee 

Convention Arrangements Committee 

Apparatus Exhibit Committee 

(e) Appointed by the Financial Vice-President 

Membership and Subscription Committee 

Sec. 5. Two Admissions Committees, one for the Atlantic Coast Section and 
one for the Pacific Coast Section, shall be appointed. The former Committee 
shall consist of a Chairman and six Fellow or Active members of the Society re- 
siding in the metropolitan area of New York, of whom at least four shall be mem- 
bers of the Board of Governors. 

The latter Committee shall consist of a Chairman and four Fellow or Active 
members of the Society residing hi the Pacific Coast area, of whom at least three 
shall be members of the Board of Governors. 

By-Law V 

Meetings 

Sec. 1. The location of each meeting of the Society shall be determined by the 
Board of Governors. 

Sec. 2. Only Honorary members, Fellows, and Active members shall be en- 
titled to vote. 

Sec. 3. A quorum of the Society shall consist in number of one-fifteenth of 
the total number of Honorary members, Fellows, and Active members as listed 
in the Society's records at the close of the last fiscal year. 

Sec. 4. The fall convention shall be the annual meeting. 

Sec. 5. Special meetings may be called by the President and upon the request 
of any three members of the Board of Governors not including the President. 

Sec. 6. All members of the Society in any grade shall have the privilege of dis- 
cussing technical material presented before the Society or its Sections. 



April, 1946 CONSTITUTION AND BY-LAWS 327 

By-Law VI 

Duties of Officers 

Sec. 1. The President shall preside at all business meetings of the Society and 
shall perform the duties pertaining to that office. As such he shall be the chief 
executive of the Society, to whom all other officers shall report, 

Sec. 2. In the absence of the President, the officer next in order as listed in 
Article IV of the Constitution shall preside at meetings and perform the duties of 
the President. 

Sec. 3. The five Vice-Presidents shall perform the duties separately enumerated 
below for each office, or as defined by the President: 

(a) The Executive Vice-President shall represent the President in such geo- 
graphical areas of the United States as shall be determined by the Board of 
Governors and shall be responsible for the supervision of the general affairs of the 
Society in such areas, as directed by the President of the Society. Should the 
President or Executive Vice-President remove his residence from the geographical 
area (Atlantic Coast or Pacific Coast) of the United States in which he resided at 
the time of his election, the office of Executive Vice-President shall immediately 
become vacant and a new Executive Vice-President elected by the Board of 
Governors for the unexpired portion of the term, the new Executive Vice-President 
to be a resident of that part of the United States from which the President or 
Executive Vice-President has just moved. 

(ft) The Engineering Vice-President shall appoint all technical committees. He 
shall be responsible for the general initiation, supervision, and coordination of 
the work in and among these committees. He may act as Chairman of any com- 
mittee or otherwise be a member ex-officio. 

(c) The Editorial Vice-President shall be responsible for the publication of the 
Society's JOURNAL and all other technical publications. He shall pass upon the 
suitability of the material for publication, and shall cause material suitable for 
publication to be solicited as may be needed. He shall appoint a Papers Com- 
mittee and an Editorial Committee. He may act as Chairman of any committee 
or otherwise be a member ex-officio. 

(d) The Financial Vice-President shall be responsible for the financial opera- 
tions of the Society, and shall conduct them in accordance with budgets approved 
by the Board of Governors. He shall study the costs of operation and the income 
possibilities to the end that the greatest service may be rendered to the members 
of the Society within the available funds. He shall submit proposed budgets to 
the Board. He shall appoint at his discretion a Ways and Means Committee, a 
Membership Committee, a Commercial Advertising Committee, and such other 
committees within the scope of his work as may be needed. He may act as Chair- 
man of any of these committees or otherwise be a member ex-officio. 

(0) The Convention Vice-President shall be responsible for the national con- 
ventions of the Society. He shall appoint a Convention Arrangements Com- 
mittee, an Apparatus Exhibit Committee, and a Publicity Committee. He may 
act as Chairman of any committee, or otherwise be a member ex-officio. 

Sec. 4. The Secretary shall keep a record of all meetings; he shall conduct the 
correspondence relating to his office, and shall have the care and custody of 
records, and the seal of the Society. 

Sec. 5. The Treasurer shall have charge of the funds of the Society and dis- 
burse them as and when authorized by the Financial Vice-President. He shall 



328 CONSTITUTION AND BY-LAWS Vol 46, No. 4 

make an annual report, duly audited, to the Society, and a report at such other 
times as may be requested. He shall be bonded in an amount to be determined 
by the Board of Governors and his bond filed with the Secretary. 

Sec. 6. Each officer of the Society, upon the expiration of his term of office, 
shall transmit to his successor a memorandum outlining the duties and policies 
of his office. 

By-Law VII 

Elections 

Sec. 1. All officers and governors shall be elected to their respective offices 
by a majority of ballots cast by the Active, Fellow, and Honorary members in the 
following manner : 

Not less than three months prior to the annual fall convention, the Board of 
Governors shall nominate for each vacancy several suitable candidates. Nomi- 
nations shall first be presented by a Nominating Committee appointed by the 
President, consisting of nine members, including a Chairman. The committee 
shall be made up of two Past-Presidents, three members of the Board of Governors 
not up for election, and four other Active, Fellow, or Honorary members, not 
currently officers or governors of the Society. Nominations shall be made by 
three-quarters affirmative vote of the total Nominating Committee. Such nomi- 
nations shall be final unless any nominee is rejected by a three-quarters vote of 
the Board of Governors present and voting. 

The Secretary shall then notify these candidates of their nomination. From 
the list of acceptances, not more than two names for each vacancy shall be se- 
lected by the Board of Governors and placed on a letter ballot. A blank space 
shall be provided on this letter ballot under each office, in which space the names 
of any Active, Fellow, or Honorary members other than those suggested by the 
Board of Governors may be voted for. The balloting shall then take place. 

The ballot shall be enclosed in a blank envelope which is enclosed in an outer 
envelope bearing the Secretary's address and a space for the member's name and 
address. One of these shall be mailed to each Active, Fellow, and Honorary 
member of the Society, not less than forty days in advance of the annual fall con- 
vention. 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the Secretary, 
sign his name and address on the latter, and mail it in accordance with the in- 
structions printed on the ballot. No marks of any kind except those above pre- 
scribed shall be placed upon the ballots or envelopes. Voting shall close seven 
days before the opening session of the annual fall convention. 

The sealed envelope shall be delivered by the Secretary to a Committee of 
Tellers appointed by the President at the annual fall convention. This com- 
mittee shall then examine the return envelopes, open and count the ballots, and 
announce the results of the election. 

The newly elected officers and governors of the general Society shall take office 
on January 1st following their election. 

By-Law VIII 
Dues and Indebtedness 

Sec. 1. The annual dues shall be fifteen dollars ($15) for Fellows and Active 
members, seven dollars and fifty cents ($7.50) for Associate members, and three 



April, 1946 CONSTITUTION AND BY-LAWS 329 

dollars ($3.00) for Student members, payable on or before January 1st of each year. 
Current or first year's dues for new members in any calendar year shall be at the 
full annual rate for those notified of acceptance in the Society on or before June 
30th; one-half the annual rate for those notified of acceptance in the Society on or 
after July 1st. 

Sec. 2. (a) Transfer of membership to a higher grade may be made at any 
time. If the transfer is made on or before June 30th the annual dues of the 
higher grade is required. If the transfer is made on or after July 1st and the 
member's dues for the full year has been paid, one-half of the annual dues of the 
higher grade is payable less one-half the annual dues of the lower grade. 

(&) No credit shall be given for annual dues in a membership transfer from a 
higher to a lower grade, and such transfers shall take place on January 1st of each 
year. 

(/;) The Board of Governors upon their own initiative and without a transfer 
application may elect, by the approval of at least three-fourths of the Board, 
any Associate or Active member for transfer to any higher grade of membership. 

Sec. 3. Annual dues shall be paid in advance. A new member who has not 
paid dues in advance shall be notified of admittance but shall not receive the 
JOURNAL and is not in good standing until initial dues are paid. All Honorary 
members, Fellows, and Active members in good standing, as defined in Section 5, 
may vote or otherwise participate in the meetings. 

Sec. 4. Members shall be considered delinquent whose annual dues for the 
year remain unpaid on February 1st. The first notice of delinquency shall be 
mailed February 1st. The second notice of delinquency shall be mailed, if neces- 
sary, on March 1st, and shall include a statement that the member's name will be 
removed from the mailing list for the JOURNAL and other publications of the 
Society before the mailing of the April issue of the JOURNAL. Members who are 
in arrears of dues on June 1st, after two notices of such delinquency have been 
mailed to then- last address of record, shall be notified their names have been re- 
moved from the mailing list and shall be warned unless remittance is received on or 
before August 1st, their names shall be submitted to the Board of Governors for 
action at the next meeting. Back issues of the JOURNAL shall be sent, if available, 
to members whose dues have been paid prior to August 1st. 

Sec. 5. (a) Members whose dues remain unpaid on October 1st may be dropped 
from the rolls of the Society by majority vote and action of the Board, or the 
Board may take such action as it sees fit. 

(b) Anyone who has been dropped from the rolls of the Society for nonpay- 
ment of dues shall, in the event of his application for reinstatement, be considered 
as a new member. 

(c) Any member may be suspended or expelled for cause by a majority vote of 
the entire Board of Governors; provided he shall be given notice and a copy in 
writing of the charges preferred against him, and shall be afforded opportunity 
to be heard ten days prior to such action. 

Sec. 6. The provisions of Sections 1 to 4, inclusive, of this By-Law VIII given 
above may be modified or rescinded by action of the Board of Governors. 

By-Law IX 

Emblem 
Sec. 1. The emblem of the Society shall be a facsimile of a four-hole film reel 



330 CONSTITUTION AND BY-LAWS Vol 46, No. 4 

with the letter 5 in the upper center opening, and the letters M, P, and E, in the 
three lower openings, respectively. The Society's emblem may be worn by 
members only. 

By-Law X 

Publications 

Sec. 1. Papers read at meetings or submitted at other times, and all material 
of general interest shall be submitted to the Editorial Board, and those deemed 
worthy of permanent record shall be printed in the JOURNAL. A copy of each 
issue shall be mailed to each member in good standing to his last address of record. 
Extra copies of the JOURNAL shall be printed for general distribution and may be 
obtained from the General Office on payment of a fee fixed by the Board of 
Governors. 

By-Law XI 

Local Sections 

Sec. 1. Sections of the Society may be authorized in any state or locality where 
the Active, Fellow, and Honorary membership exceeds 20. The geographic 
boundaries of each Section shall be determined by the Board of Governors. 

Upon written petition, signed by 20 or more Active members, Fellows, and Hon- 
orary members, for the authorization of a Section of the Society, the Board of 
Governors may grant such authorization. 

Section Membership 

Sec. 2. All members of the Society of Motion Picture Engineers in good stand- 
ing residing in that portion of any country set apart by the Board of Governors 
tributary to any local Section shall be eligible for membership in that Section, and 
when so enrolled they shall be entitled to all privileges that such local Section may, 
under the General Society's Constitution and By-Laws, provide. 

Any member of the Society in good standing shall be eligible for nonresident 
affiliated membership of any Section under conditions and obligations prescribed 
for the Section. An affiliated member shall receive all notices and publications of 
the Section but he shall not be entitled to vote at sectional meetings. 

Sec. 3. Should the enrolled Active, Fellow, and Honorary membership of a 
Section fall below 20, or should the technical quality of the presented papers fall 
below an acceptable level, or the average attendance at meetings not warrant the 
expense of maintaining the organization, the Board of Governors may cancel its 
authorization. 

Section Officers 

Sec. 4. The officers of each Section shall be a Chairman and a Secretary- 
Treasurer. The Section chairmen shall automatically become members of the 
Board of Governors of the General Society, and continue in such positions for the 
duration of their terms as chairmen of the local Sections. Each Section officer 
shall hold office for one year, or until his successor is chosen. 

Section Board of Managers 

S ec 5._Tne Board of Managers shall consist of the Section Chan-man, the 
Section Past-Chairman, the Section Secretary-Treasurer, and six Active, Fellow, or 



April, 1946 CONSTITUTION AND BY-LAWS 331 

Honorary members. Each manager of a Section shall hold office for two years, 
or until his successor is chosen. 

Section Elections 

Sec. 6. The officers and managers of a Section shall be Active, Fellow, or 
Honorary members of the General Society. All officers and managers shall be 
elected to their respective offices by a majority of ballots cast by the Active, Fel- 
low, and Honorary members residing in the geographical area covered by the 
Section. 

Not less than three months prior to the annual fall convention of the Society, 
nominations shall be 'presented to the Board of Managers of the Section by a 
Nominating Committee appointed by the Chairman of the Section, consisting of 
seven members, including a chairman. The Committee shall be composed of the 
present Chairman, the Past-Chairman, two other members of the Board of Man- 
agers not up for election, and three other Active, Fellow, or Honorary members of 
the Section not currently officers or managers of the Section. Nominations shall 
be made by a three-quarters affirmative vote of the total Nominating Committee. 
Such nominations shall be final, unless any nominee is rejected by a three-quarters 
vote of the Board of Managers, and in the event of such rejection the Board of 
Managers will make its own nomination. 

The Chairman of the Section shall then notify these candidates of their nomi- 
nation. From the list of acceptances, not more than two names for each vacancy 
shall be selected by the Board of Managers and placed on a letter ballot. A blank 
space shall be provided on this letter ballot under each office, in which space the 
names of any Active, Fellow, or Honorary members other than those suggested 
by the. Board of Managers may be voted for. The balloting shall then take place. 

The ballot shall be enclosed in a blank envelope which is enclosed in an outer 
envelope bearing the local Secretary-Treasurer's address and a space for the 
member's name and address. One of these shall be mailed to each Active, Fellow, 
and Honorary member of the Society residing hi the geographical area covered by 
the Section, not less than forty days in advance of the annual fall convention. 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the Secretary- 
Treasurer, sign his name and address on the latter, and mail it in accordance with 
the instructions printed on the ballot. No marks of any kind except those above 
prescribed shall be placed upon the ballots or envelopes. Voting shall close seven 
days before the opening session of the annual fall convention. 

The sealed envelopes shall be delivered by the Secretary-Treasurer to his 
Board of Managers at a duly called meeting. The Board of Managers shall then 
examine the return envelopes, open and count the baUots, and announce the 
results of the election. 

The newly elected officers and managers shall take office on January 1st follow- 
ing their election. 

Section Business 

Sec. 7. The business of a Section shall be conducted by the Board of Managers. 

Section Expenses 

Sec. 8. (a) As early as possible in the fiscal year, the Secretary-Treasurer of 
each Section shall submit to the Board of Governors of the Society a budget of 
expenses for the year. 

(5) The Treasurer of the General Society may deposit with each Section Secre- 



332 CONSTITUTION AND BY-LAWS Vol 46, No. 4 

tary-Treasurer a sum of money, the amount to be fixed by the Board of Governors, 
for current expenses. 

(c) The Secretary-Treasurer of each Section shall send to the Treasurer of the 
General Society, quarterly or on demand, an itemized account of all expenditures 
incurred during the preceding interval. 

(d) Expenses other than those enumerated hi the budget, as approved by the 
Board of Governors of the General Society, shall not be payable from the general 
funds of the Society without express permission from the Board of Governors. 

(e) A Section Board of Managers shall defray all expenses of the Section not 
provided for by the Board of Governors, from funds raised locally by donation, 
or fixed annual dues, or by both. 

(/) The Secretary of the General Society shall, unless otherwise arranged, supply 
to each Section all stationery and printing necessary for the conduct of its business. 

Section Meetings 

Sec. 9. The regular meetings of a Section shall be held in such places and at 
such hours as the Board of Managers may designate. 

The Secretary-Treasurer of each Section shall forward to the Secretary of the 
General Society, not later than five days after a meeting of a Section, a statement 
of the attendance and of the business transacted. 

Section Papers 

Sec. 10. Papers shall be approved by the Section's Papers Committee previ- 
ously to their being presented before a Section. Manuscripts of papers presented 
before a Section, together with a report of the discussions and the proceedings of 
the Section meetings, shall be forwarded promptly by the Section Secretary- 
Treasurer to the Secretary of the General Society. Such material may, at the dis- 
cretion of the Board of Editors of the General Society, be printed in the Society's 
publications. 

Constitution and By-Laws 

Sec. 11. Sections shall abide by the Constitution and By-Laws of the Society 
and conform to the regulations of the Board of Governors. The conduct of Sec- 
tions shall always be in conformity with the general policy of the Society as fixed 
by the Board of Governors. 

By-Law XII 

Amendments 

Sec. 1 . These By-Laws may be amended at any regular meeting of the Society 
by the affirmative vote of two-thirds of the members present at a meeting who 
are eligible to vote thereon, a quorum being present, either on the recommendation 
of the Board of Governors or by a recommendation to the Board of Governors 
signed by any ten members of Active or higher grade, provided that the proposed 
amendment or amendments shall have been published in the JOURNAL of the 
Society, hi the issue next preceding the date of the stated business meeting of the 
Society at which the amendment or amendments are to be acted upon. 

Sec. 2. In the event that no quorum of the voting members is present at the 
time of the meeting referred to in Section 1, the amendment or amendments shall 



April, 1946 JOURNAL AWARD AND PROGRESS MEDAL AWARD 333 

be referred for action to the Board of Governors. The proposed amendment or 
amendments then become a part of the By-Laws upon receiving the affirmative 
vote of three-quarters of the Board of Governors. 



JOURNAL AWARD AND PROGRESS MEDAL AWARD 



In accordance with the provisions of Administrative Practices of the So- 
ciety, the regulations of procedure for the Journal Award and the Progress 
Medal Award, a list of the names of previous recipients, and the reasons therefor, 
shall be published annually in the JOURNAL as follows: 

JOURNAL AWARD 

The Journal Award Committee shall consist of five Fellows or Active members 
of the Society, appointed by the President and confirmed by the Board of Gover- 
nors. The Chairman of the Committee shall be designated by the President. 

At the fall convention of the Society a Journal Award Certificate shall be pre- 
sented to the author or to each of the authors of the most outstanding paper 
originally published in the JOURNAL of the Society during the preceding calendar 
year. 

Other papers published in the JOURNAL of the Society may be cited for Honorable 
Mention at the option of the Committee, but in any case should not exceed five in 
number. 

The Journal Award shall be made on the basis of the following qualifications : 

(2) The paper must deal with some technical phase of motion picture engineer- 
ing. 

(2} No paper given in connection with the receipt of any other Award of the 
Society shall be eligible. 

(5) In judging of the merits of the paper, three qualities shall be considered, 
with the weights here indicated : 

(a) Technical merit and importance of material 45 per cent. 

(6) Originality and breadth of interest 35 per cent. 

(c) Excellence of presentation of the material 20 per cent. 

A majority vote of the entire Committee shall be required for the election to the 
Award. Absent members may vote in writing. 

The report of the Committee shall be presented to the Board of Governors at 
their July meeting for ratification. 

These regulations, a list of the names of those who have previously received the 
Journal Award, the year of each Award, and the titles of the papers shall be pub- 
lished annually in the April issue of the JOURNAL of the Society. In addition, the 
list of papers selected for Honorable Mention shall be published in the JOURNAL of 
the Society during the year current with the Award. 



334 JOURNAL AWARD AND PROGRESS MEDAL AWARD Vol 46, No. 4 

The Awards in previous years have been as follows: 

1934 P. A. Snell, for his paper entitled "An Introduction to the Experi- 
mental Study of Visual Fatigue." (Published May, 1933.) 

1935-yL. A. Jones and J. H. Webb, for their paper entitled "Reciprocity 
Law Failure in Photographic Exposure." (Published Sept., 1934.) 

1936 E. W. Kellogg, for his paper entitled "A Comparison of Variable- 
Density and Variable- Width Systems." (Published Sept., 1935.) 

1937 D. B. Judd, for his paper entitled "Color Blindness and Anomalies of 
Vision." (Published June, 1936.) 

1938 K. S. Gibson, for his paper entitled "The Analysis and Specification of 
Color." (Published Apr., 1937.) 

1939 H. T. Kalmus, for his paper entitled "Technicolor Adventures in 
Cinemaland." (Published Dec., 1938.) 

1940 R. R. McNath, for his paper entitled "The Surface of the Nearest 
Star." (Published Mar., 1939.) 

1941 J. G. Frayne and Vincent Pagliarulo, for their paper entitled "The 
Effects of Ultraviolet Light on Variable-Density Recording and Printing." 
(Published June, 1940.) 

1942 W. J. Albersheim and Donald MacKenzie, for their paper entitled 
"Analysis of Sound-Film Drives." (Published July, 1941.) 

1943 R. R. Scoville and W. L. Bell, for their paper entitled "Design and 
Use of Noise-Reduction Bias Systems." (Published Feb., 1942; Award made 
Apr., 1944.) 

1944 J, I. Crabtree, G. T. Eaton, and M. E. Muehler, for their paper en- 
titled "Removal of Hypo and Silver Salts from Photographic Materials as 
Affected by the Composition of the Processing Solutions." (Published July, 

1945 C. J. Kunz, H. E. Goldberg, and C. E. Ives, for their paper entitled 
"Improvement in Illumination Efficiency of Motion Picture Printers." (Pub- 
lished May, 1944.) 

The present Chairman of the Journal Award Committee is F. E. Carlson. 



PROGRESS MEDAL AWARD 

The Progress Medal Award Committee shall consist of five Fellows or Active 
members of the Society, appointed by the President and confirmed by the Board 
of Governors. The Chairman of the Committee shall be designated by the 
President. 

The Progress Medal may be awarded each year to an individual in recognition 
of any invention, research, or development which, in the opinion of the Com- 
mittee, shall have resulted in a significant advance in the development of motion 
picture technology. 

Any member of the Society may recommend persons deemed worthy of the 
Award. The recommendation in each case shall be in writing and in detail as to 
the accomplishments which are thought to justify consideration. The recom- 
mendation shall be seconded in writing by any two Fellows or Active members 
of the Society, who shall set forth then- knowledge of the accomplishments of the 
candidate which, in their opinion, justify consideration. 

A majority vote of the entire Committee shall be required to constitute an 
Award of the Progress Medal. Absent members may vote in writing. 

The report of the Committee shall be presented to the Board of Governors 
at their July meeting for ratification. 



April, 1946 JOURNAL AWARD AND PROGRESS MEDAL AWARD 335 

The recipient of the Progress Medal shall be asked to present a photograph of 
himself to the Society and, at the discretion of the Committee, may be asked to 
prepare a paper for publication in the JOURNAL of the Society. 

These regulations, a list of the names of those who have previously received 
the Medal, the year of each Award, and a statement of the reason for the Award 
shall be published annually in the April issue of the JOURNAL of the Society. 

Previous Awards have been as follows : 

The 1935 Award was made to E. C. Wente, for his work in the field of sound 
recording and reproduction. (Citation published Dec., 1935.) 

The 1936 Award was made to C. E. K. Mees, for his work in photography. 
(Citation published Dec., 1936.) 

The 1937 Award was made to E. W. Kellogg, for his work in the field of sound 
reproduction. (Citation published Dec., 1937.) 

The 1938 Award was made to H. T. Kalmus, for his work in developing color 
motion pictures. (Citation published Dec., 1938.) 

The 1939 Award was made to L. A. Jones, for his scientific researches in the 
field of photography. (Citation published Dec., 1939.) 

The 1940 Awar*d was made to Walt Disney, for his contributions to motion 
picture photography and sound recording of feature and short cartoon films 
(Citation published Dec., 1940.) 

The 1941 Award was made to G. L. Dimmick, for his development activities 
in motion picture sound recording. (Citation published Dec., 1941.) 

No Awards were made in 1942 and 1943. 

The 1944 Award was made to J. G. Capstaff, for his research and develop- 
ment of films and apparatus used in amateur cinematography. (Citation pub- 
lished Jan., 1945.) 
No AWard was made in 1945. 

The present Chairman of the Progress Medal Award Committee is E. A. Willi- 
ford. 



REPORT OF THE TREASURER 



SOCIETY OF MOTION PICTURE ENGINEERS 
JANUARY 1-DECEMBER 31, 1945 

Members' Equity, Jan. 1, 1945: $40,855.75 

Receipts, Jan -Dec., 1945: 

Membership Dues $17,813.92 

Sustaining Memberships 8,087 . 50 

Publications (Subscriptions, Re- 
prints, Journals, Book, Standards, . 
etc.) 7,576.43 
Test Films 79,615.87 
Other (Conferences, Interest, etc.) 838.87 

Total Receipts $1 13,932 . 59 

Disbursements, Jan.-Dec., 1945: 

General Office (Salaries, Rent, Sup- 
plies, Tel. and Tel., Equipment, 
Travel, Postage, etc.) 

Publications (Journal, Reprints, etc.) 

Test Films 

Dues and Fees to Other Organiza- 
tions (ASA, RTPB, NFPA, ISCC) 

Sections (Atlantic and Pacific) 

Other (Committees, Awards, etc.) 

Total Disbursements 78,886 . 98 

Excess Receipts Over Disbursements, 1945 35,045.61 

Members' Equity, Dec. 31, 1945 $75,901 . 36 

* Subject to Renegotiation 

Respectfully submitted, 
EARL I. SPONABLE, Treasurer 

The cash records of the Treasurer were audited for the year ended December 31, 
1945, by Sparrow, Waymouth and Company, certified public accountants, New 
York, and are in conformity with the above report. 

M. R. BOYER, 

Financial Vice-P resident 



15,798.58 

9,641.69 

51,502.44 

835.00 
590.36 
518.91 



336 



SOCIETY ANNOUNCEMENTS 

ATLANTIC COAST SECTION MEETING 

Foreign language releases of American motion picture films were discussed by 
W. A. Pozner of MGM International Films Corporation, New York, at the meet- 
ing of the Atlantic Coast Section of the Society on February 13. The meeting was 
held in the auditorium of the Museum of Modern Art, New York. 

Mr. Pozner reviewed the history of dubbing methods, discussed the importance 
of sound perspective, and gave a detailed description of the dubbing technique 
used by MGM. The talk was followed by 3 motion picture shorts illustrating 
the special projection magazine and the MGM dubbing process, an example of 
an original English release print with alternate Spanish and English sound tracks, 
and an original foreign film with English sound track. 

Because of the wide interest in the industry in preparing foreign versions of en- 
tertainment films, a large audience turned out for this presentation. MGM Inter- 
national night was one of a series of meetings arranged by the Atlantic Coast Section 
to promote wider knowledge of industry techniques and practices. The paper will 
be published in a forthcoming issue of the JOURNAL. 

PACIFIC COAST SECTION MEETING 

Stuart W. Seeley, director of the Industry Service Laboratory of Radio Corpo- 
ration of America, New York, described the latest developments in television and 
their relationship to the motion picture industry before members and guests of 
the Pacific Coast Section of the Society in Hollywood, February 20. 

Mr. Seeley also gave a highly interesting description of the "Shoran" system 
which made possible accurate blind bombing through ability to locate the plane's 
position in the air to an accuracy of a few feet at all times. The current applica- 
tion of Shoran to mapping was also explained. 

The Walt Disney Studios was host to the gathering of some 250 members, stu- 
dio executives, and others. 

NOMINATIONS FOR ANNUAL ELECTIONS 

In accordance with Administrative Practices of the Society, a Committee on 
Nominations has been appointed by the President to recommend candidates for 
offices expiring December 31, 1946. General elections are held prior to the Octo- 
ber Technical Conference. The offices expiring and incumbents are given on the 
reverse of the contents page of this issue. 

Voting members of the Society (Honorary, Fellow, and Active) are invited to 
submit recommendations for candidates to the Nominating Committee. Only 
Honorary, Fellow, and Active members may hold office. Names should be sent to 
E. M. Honan, Chairman of the Committee, 6601 Romaine St., Hollywood 38, 
Calif., or to any committee member listed on page 317 of this issue. A report will 
be submitted to the Board of Governors at the July meeting. 

337' 



338 SOCIETY ANNOUNCEMENTS Vol 46, No. 4 

EMPLOYMENT SERVICE 

In the December 1944 JOURNAL the Society announced the inauguration of a 
free Employment Service to assist members in making business connections, 
and to offer industry organizations an opportunity to place their engineering 
personnel requirements before a large group of experienced engineers and special- 
ists in the motion picture industry. This mutual service has been carried in 
the JOURNAL since, showing both Positions Open and Positions Wanted. 

The General Office has received letters from individuals and companies com- 
mending the service, and many members have made new business affiliations 
through this medium. The following excerpt from a letter by a returned war 
veteran-member is typical : 

"Happy to report my ad in 'Positions Wanted' section of the SMPE 
Journal of January and February, 1946, has had results and I have gained 
a satisfactory position. . . . My sincere thanks to the staff for the inser- 
tion of this ad " 

The Society believes that this medium may be helpful to other members, es- 
pecially veterans, and companies in the motion picture industry and it is re- 
quested that any interested member or company send particulars to the General 
Office for insertion in the next JOURNAL. Top experts in the field read the 
JOURNAL regularly and there are excellent prospects of obtaining the position 
or engineer desired. The Society, however, reserves the right both to edit or 
reject any notice submitted for publication. 

POSITIONS OPEN 

Designer and engineer experienced in optics, lighting, and microphotog- 
raphy, capable of designing microfilm reading equipment and products 
related to microfilm industry. Reply to Microstat Corporation, 18 
West 48th St., New York 19, N.Y. 



Position available for Optical Designer, capable of handling the calcula- 
tion and correction of aberrations in photographic and projection lens 
systems. Junior designers or engineers will be considered. Write 
fully giving education, experience, and other qualifications to Director 
of Personnel, Bell and Howell Company, 7100 McCormick Road, Chi- 
cago 45, 111. 



Motion picture studio in Bombay, India, has positions open for profes- 
sional motion picture camerman with studio and location experience; 
sound recording engineer experienced- in installation, maintenance and 
operation of recording equipment; motion picture processing labora- 
tory supervisor; and professional make-up artist. Five-year contracts 
at favorable terms are offered to those qualified. Write or cable direct 
to Personnel Manager, Dawlat Corporation Ltd., Patel Chambers, French 
Bridge, Bombay 7, India, giving experience, etc., in detail. 



New film production unit to be located at Athens, Georgia, needs film 
editor-writer and film director. Experience in 16-mm as well as 35-mm 
production desirable. Southern background or interest in South pre- 



April, 1946 SOCIETY ANNOUNCEMENTS 339 

f erred but not essential. Write giving full details of experience, etc., to 
Nicholas Read, The National Film Board, Ottawa, Canada. 

POSITIONS WANTED 

Sound recording engineer, 16- or 35-mm equipment, studio or location 
work, single or double system. Free to travel. For details write T. J. K., 
354 Ninth Ave., New York 1, N.Y. 

Honorably discharged veteran with 15 years' experience in all phases of 
motion picture production, including film editing, directing, producing. For 
details write F. A., 30-71 34th St., Long Island City 3, N.Y. Telephone 
AStoria 8-0714. 



Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
den Station, Baltimore 11, Maryland. 



Honorably discharged veteran with 10 years' experience in projection 
and installation of projection and sound equipment, both for booth and 
back-stage. Prefer to locate in California, Oregon or Nevada. For ad- 
ditional details write F.A.N., Box 113, Holley, Oregon. 



MEMBERS OF THE SOCIETY 

LOST IN THE SERVICE OF 

THEIR COUNTRY 



FRANKLIN C. GILBERT 



ISRAEL H. TILLES 



MORGAN L. HOBART 



HARRY B. CUTHBERTSON 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 MAY, 1946 No. 5 

CONTENTS 

PAGE 

A Lens Calibrating System C. R. DAILY 343 

Ansco Color for Professional Motion Pictures 

H. H. DUERR AND H. C. HARSH 357 

Sensitometric Evaluation of Reversible Color Film 

R. H. BINGHAM 368 

A Phototube for Dye Image Sound Track 

A. M. GLOVER AND A. R. MOORE 379 

Preliminary Sound Recording Tests with Variable- 
Area Dye Tracks R. O. DREW AND S. W. JOHNSON 387 

Behavior of a New Blue-Sensitive Phototube in Theater 

Sound Equipment J. D. PHYFE 405 

Electronic Shutter Testers R. F. REDEMSKE 409 

Book Reviews 424 

Current Literature 426 

Society Announcements 428 



Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL, must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCIETY of MOTION PICTURE ENGINEERS 

MOTL PENNSYLVANIA NCW YOftKi, N-Y- TCL. PSNN. 6 O62O 

HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 
*President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
*Past-President: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
*Executive Vice-President: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
** Engineering Vice-President: JOHN A. MAURER, 

37-01 31st St., Long Island City 1, N. Y. 
* Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
** Financial Vice-President: M. R. BOYER, 

350 Fifth Ave., New York 1. 
* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
*Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
^Treasurer: EARL I. SPONABLE, 

460 West 54th St., New York 19. 

Governors 

"fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
** WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
*JHOLLIS W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairtnan, Atlantic Coast Section. 
**Term expires December 31, 1947. tChairman, Pacific Coast Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol46 MAY, 1946 No. 5 

A LENS CALIBRATING SYSTEM * 
C. R. DAILY** 



Summary. This paper describes a method for the calibration of the effective 
f/stop value of a camera lens in terms of the light transmitted by the lens. Owing to 
light losses within a lens, the f /value determined in this manner will normally be 
numerically larger than the values obtained from the geometry of the lens. However, 
stop values based on transmission are of considerably greater value to a cameraman 
than values based on apertures. 

An interrupted, cdlimated beam of light falls on the entrance port of an integrating 
sphere. The lens to be calibrated is centered over the opening and the light output from 
the sphere measured through a stable a-c amplifier equipment. A calibration curve 
is then obtained for the equipment wherein metal plates with holes of known diameters 
are placed over the entrance of the sphere in place of the lens. The effective aperture 
of the lens is then defined as the size of opening which, will pass the same total amount 
of light as the lens under calibration. This method avoids the use of secondary stand- 
ards and should be capable of duplication in any laboratory. 

The light transmission properties of a photographic objective lens 
are materially improved if the lens surfaces are coated to reduce re- 
flection losses. Therefore the true photographic speed of a lens is not 
given directly by the effective aperture ratio and a cameraman must 
make allowances for variations from the marked //values on various 
types of lenses in order to obtain a correct exposure, since the marked 
//values do not include transmission losses. The error increases with 
the number of air-glass surfaces, and the effect of coating a lens should 
be to reduce the difference between the transmission-effective and the 
geometric-effective apertures. This variable, coupled with differ- 
ences in original factory calibration, has led a number of investiga- 
tors to study the problem and as a result several methods for cali- 
brating lenses in terms of the true photographic speed have been put 
into use. 

* Presented Oct. 16, 1945, at the Technical Conference in New York. ' 
"* Paramount Pictures, Inc., Hollywood. 

343 



344 C. R. DAILY Vol 46, No. 5 

The method of lens calibration described in this paper was de- 
veloped to provide a standard which could be reproduced in any labo- 
ratory and which would be photographically correct for the central 
part of the image. The use of fixed aperture stops of known diameter 
satisfied the first requirement for standard reference apertures, while 
the use of collimated light simulated a point source of light at infinity 
and therefore should give results equivalent to the image intensity in 
the center of the field. Close correlation to photographic density 
tests has been obtained which appears to validate the method, 

In earlier work on lens calibration by Technicolor and Twentieth 
Century-Fox, 1 the lenses to be tested were placed in front of a large 
illuminated ground glass. The cone of light collected by and passing 
through the lens was focused on a standard motion picture aperture 
plate. A phototube and suitable meters. then measured the total 
light falling on the phototube behind the aperture, and the ratio of 
the light transmitted served as the basis for the lens diaphragm cali- 
bration. Such methods, however, are not absolute and it is necessary 
to refer all lenses to one which is retained as a "standard." Likewise, 
a large cone of light is picked up and since the meter gives an average 
reading for the total light transmitted, it will not be truly propor- 
tional to the light in the center of the field. This "fall-off" effect is 
more noticeable in the case of short focus lenses. 

More recently, 2 the Signal Corps Photographic Center at Long 
Island City, New York, has been studying this problem and described 
the preliminary form of their apparatus before the Society in Holly- 
wood in May, 1945. They undertook this same problem of recali- 
bration because of the wide variation in exposure which was being 
observed among the various lenses used by them. 

Work on this same problem was undertaken several years ago at 
Paramount, 3 and the first results were reported in August, 1942. 
Following the interruption of this study because of the war, the prob- 
lem was again opened for study and a new method developed which 
should materially increase the accuracy, reproducibility, and speed 
with which a lens can be calibrated, and at the same time be consid- 
ered as standard. 

Since the f /number of a lens is defined on the basis of parallel rays, 
the electrical f/ , or t/ number as it uill be called here, obtained by the 
proposed method should be equal to the f /value assigned to a lens by 
classical methods of calibration, corrected for the axial transmission 
losses of the lens. 



May, 1946 



A LENS CALIBRATING SYSTEM 



345 



Proposed Calibration System. Optical Schematic. Fig. 1 shows 
a schematic of the optical portion of the calibrating equipment. A 
3-in. diameter, 4V2-in. focal length condenser lens images the biplane 
filament of a 1000-w projection lamp on a 1 /2-in. diameter round aper- 
ture in a metal plate. In order to more evenly illuminate this aper- 
ture, a diffusing glass which is ground on both sides is placed over the 
aperture. The filament image at this point is approximately 3 /4 in. 
sq. An//3.5, 30-cm focal length lens, located at a distance equal to 
its focal length from the J /2-in. aperture, directs a homogeneous colli- 
mated beam of light toward the opening in the integrating sphere. 
This collimated beam does not vary appreciably in brightness either 
along the beam or across a 1.8-in. diameter at the center of the 3-in. 
diameter beam. 



SPHERICAL MIRROR 

\ DIFFUSION DISC 



FIXED STOP OR LENS 
TO BE CALIBRATED 



PHOTOCELL 




f CONDENSER LENS 

/PROJECTS LAMP /STOP ,NTEGRAT,NG SPHERE 

FIG. 1. Optical schematic of lens calibrating equipment. 



A light interrupter wheel is mounted near the V2-m. aperture to 
provide a substantially sinusoidal, 400-cycle per sec interruption of 
the light beam in order to permit the use of stabilized a-c amplifiers 
in the measuring system. The light is picked up by means of an in- 
tegrating sphere which has a 12-in. internal diameter. A General 
Electric type GL-441 vacuum phototube is inserted inside the sphere. 
This cell has an S-4 surface and provides a maximum sensitivity at 
540 niju when exposed to a tungsten lamp operating at 2848 K. 

Fig. 2 shows the spectral characteristic of the combined phototube 
and tungsten lamp under these conditions. The response curve ap- 
pears to be adequate for comparative tests on black-and-white film. 
A refinement of the apparatus is now under way to mount 2 photo- 
tubes with selective filters in order to obtain a substantially flat re- 
sponse from 400 to 700 mju, for a better determination of lens proper- 
ties when used with color. 



346 



C. R. DAILY 



Vol 46, No. 5 



Electrical Schematic. An a-c measuring system was adopted for 
several reasons: 

(7) Higher gain stability can be obtained with respect to power 
supply voltage variations, etc., compared with d-c amplifier systems. 

(2) Ease of providing precise attenuation to measure relative 
electrical input. 



3 w RELATIVE SENSITIVITY 


































/ 


f 


\ 










/ 






\ 






y 


/ 






\ 






/ 








\ 




/ 


/ 










\ 


/ 












\ 


400 450 500 550 600 650 7C 
WAVELENGTH IN MILLIMICRONS 

t. 2. Over-all special sensitivity of a GL-441 phototul 



and tungsten lamp operating at 2848 K. 



(3) High signal-to-noise ratio can be obtained with band-pass 
filters, when single-frequency amplification only is being used. 

The schematic of the electrical system is shown in Fig. 3. The 
phototube is resistance-coupled to a 2-stage phototube amplifier 
which has a 14-db interstage gain switch that is cut in when measur- 
ing with the largest diaphragm openings, or the largest lenses, to pre- 
vent overload of the second stage. A vernier gain control is also pro- 
vided to permit accurate adjustment of the output for reference cali- 
bration with a given fixed stop. This vernier control takes care of 
variations in phototube sensitivity and aging of the lamp. 



May, 1946 



A LENS CALIBRATING SYSTEM 



347 



The phototube amplifier is followed by two 30-db variable precision 
bridged T attenuators, a 70-db gain feedback-stabilized line amplifier, 




PEC 
AMPLIFIER 


X X 


LINE 

AMPLIFIER 




400 /^ 
BAND 
PASS 
FILTER 


T~$ 






PRECISION 






ATTENUATORS 



FIG. 3. Block schematic of electrical equipment used with lens calibrating 

equipment. 

a 400-cycle band-pass filter, terminating resistance, and a copper ox- 
ide rectifier-type volume indicator. 



CJO 

10 
.9 
.8 

~ .7 
S 

i .5 

.4 
ct 

Ul 

I' 5 
1* 



x 

.1 














/ 












f 














s 














f 












2 














/ 














/ 












/ 














/ 










, 


/ 












f 
















/ 












J07 




f 












/ 














.05 
C 


/ 














10 20 30 40 50 60 70 



TOTAL ATTENUATION ( DB ) 
PHOTOCELL AMPLIFIER AND ATTENUATORS 

FIG. 4. Required attenuation versus calibrating stop 
diameter. 



Electrical Calibration of a Lens. The system is first calibrated 
by placing a series of metal plates over the entrance to the integrat- 
ing sphere. These plates have been drilled with circular holes of 
precisely known diameters so that the total area of the opening is 



348 



C. R. DAILY 



Vol 46, No. 5 



known. With the projection lamp held at constant voltage, the total 
attenuation for reference deflection of the volume indicator is read 
for the whole series of fixed stops which vary from 0.068 hi. to 1.813 
in. in diameter. 

The calibration curve obtained for such a series of fixed stops is 
shown in Fig. 4 and has a slope in decibels of 40 Iogi di/d2, where di and 
d 2 are any 2 diameters of fixed stop opening, assuming a uniform beam 




FIG. 5. Integrating sphere with a calibrating stop 
mounted over sphere opening. 



of light. It will be noted that this calibration curve is linear over the 
range covered which indicates linearity of the light beam over the 
aperture, of the collection of light by the sphere, of the phototube and 
of the amplifiers. Changes in lamp voltage cause output changes of 
approximately 0.25 db per volt, corresponding to effective lens dia- 
phragm variations of 0.045 stops per volt. 

Fig. 5 shows a photograph of the integrating sphere with one of the 
fixed calibrating stops mounted in position. In order to determine the 
uniformity of the light beam over the entrance to the sphere, a plate 



May, 1946 



A LENS CALIBRATING SYSTEM 



349 



with a very small opening was moved over the entrance hole to the 
sphere. It was found that the volume indicator deflection was essen- 
tially constant for any position of the small hole over the 1.813-in. 
diameter opening. 

Fig. 6 shows a photograph of the sphere with a lens to be calibrated 
installed over the opening. After the system has been calibrated as 
indicated in the previous paragraphs, the lens is placed over the open- 




FIG. 6. 



Integrating sphere with lens mounted over the 
sphere opening. 



ing and the electrical attenuation adjusted for reference deflection of 
the volume indicator for the various lens diaphragm positions. For 
any attenuation obtained from the lens, there is a corresponding fixed 
stop diameter which would require the same attenuation. This cor- 
relation to precisely known stop diameters is the basis of the absolute 
calibration provided by this equipment. 

Table 1 shows a typical calibration for lens A , which had a nominal 
focal length of one inch. The precise focal length as determined by 
the method described in the following section indicated that F 
1.01 in. 



350 C. R. DAILY Vol 46, No. 5 







TABLE 1 






Original //- 


Electrical 


Effective Dia- 






Value Marked 
on Lens 


Attenuation 
(db) 


phragm Opening 
Diameter (In.) 


t Corrected 
//Value 


n Diaphragm 
Marking Error 


1.8 


42.0 


0.444 


2.27 


-0.52 


2.3 


41.0 


0.435 


2.32 


-0.02 


3.2 


36.2 


0.330 


3.06 


0.13 


4.5 


29.4 


0.234 


4.32 


0.12 


6.3 


25.0 


0.173 


5.88 


0.20 


9 


20.0 


0.130 


7.77 


0.43 


12 


14.0 


0.090 


11.20 


0.20 



The electrical attenuation was adjusted as indicated above for each 
diaphragm position. The effective diameters were then read from 
the stop calibration curve shown in Fig. 4. The effective //value of 
the lens, which will be tentatively designated by t, was then calcu- 
lated from the denning equation 

_ Focal length 

Fixed stop diameter which will produce the same electrical output as the lens 

U) 

'The correlation is fairly close for this lens, although some lenses 
have been tested where the //value differed from the originally 
marked //value by as much as one stop. The last column of the 
table shows the error in fractions of a stop. The conversion to error 
of diaphragm marking can be readily made by means of the formula 



n = 6.65 logio^; or 6.65 logio 

/i n 

where n is the fractional stop difference between / 2 and t\. 

In Table 2 is shown the range of attenuations which will be en- 
countered for lenses at 1- to 4-in. focal length, at diaphragm openings 
of from //1. 8 to// 18. It will be noted that a range in attenuation of 
64 db is required and care must therefore be taken in the design and 
wiring to prevent overloading of the phototube amplifier and to insure 
that the indicated attenuation is being obtained. , 

TABLE 2 

Required Attenuation (Db) 

Focal Length of Lens (In.) 

/ 1 

1.8 

2.3 
4.5 
9 
18 



1 


2 


3 


4 


45.24 


57.28 


64.32 


69.32 


41.00 


53.04 


60.08 


65.08 


29.34 


41.38 


48.42 


53.42 


17.32 


29.36 


36.40 


41.40 


5.24 


17.28 


24.32 


29.32 



May, 1946 



A LENS CALIBRATING SYSTEM 



351 



Determination of Focal Length. In order to determine the 
//value, one may assume the focal length as given. However, if it is 
desirable to recheck this value, a direct photographic method may be 
used which is quite precise and direct. The method to be described 
was selected from several which 
are in use, since it employed 
available equipment and could 
be readily applied. 

An outdoor location near the 
laboratory was selected, a transit 
set up at a fixed point and the 
angles separating a number of 
distant reference points measured 
to an accuracy of the nearest 
minute of arc. The reference 
points were more than 200 ft 
distant and the angular separa- 
tions were selected so that it 
would be possible to use points 
for the various focal length lenses 
which would be separated on the 
image by approximately two- 
thirds of a standard motion 
picture frame width. Fig. 7 
shows the geometry of this setup. 

The lens under test was 
mounted on a Speed Graphic 
camera which had been equipped 
with quick change mounts. The 
image of the distant reference 
points was carefully centered 
and focused on the ground glass. 
A photograph was then taken on a 
glass plate and processed in fine-grain developer. After fixing, wash- 
ing, and drying the plate, the distance between the reference points 
on the photograph was measured on a precision toolmaker's micro- 
scope. The focal length was then computed from the formula 




CAMERA 



PLATE 



FIG. 7. Determination of focal length 
by a photographic method. 



where 6 is the angle subtended by the distant reference points and h is 



352 



C. R. DAILY 



Vol 46, No. 5 



the distance between the reference points on the image plate. In the 
determination of focal length by this method, the accuracy is much 
better than one per cent and is satisfactory in lieu of an optical bench 
which might not always be available. Table 3 shows the angle and 
distance relationships obtained on a number of lenses. 

Photographic Confirmation. In order to check the accuracy of 
the proposed method of //stop calibration, photographic tests were 



48 INCHES 



-WHITE SCREEN 



BROADS 



,O 



Q 




f=^ ^-MOTION PICTURE CAMERA 



FIG. 8. Method of illuminating a test screen for photo- 
graphic confirmation test with a camera. 

made using various lenses which had been calibrated by the above 
method. A white test card approximately 4 ft sq was set up and uni- 
formly illuminated by diffuse light from four 1000-w lamps, as shown 
in Fig. 8. A standard motion picture camera, driven at constant 
speed by an interlock motor was set up 4 ft from and facing the il- 
luminated test card. 

The camera was loaded with a roll of standard 35-mm picture nega- 
tive film and exposure tests were then made on the various lenses at 
several apertures each. The film was given standard machine process- 
ing to insure uniformity of development. 



May, 1946 



A LENS CALIBRATING SYSTEM 



353 



In the photography of a light source at a distance greater than 100 X 
focal length, the intensity in the center of the field of the image on 
the photographic plate should be substantially independent of the 
focal length of the lens, for a given //value. Therefore, if a photo- 




20 



FIG. 9. Photographic confirmation data. The visual 
diffuse density of the film as measured in the center of the 
film area is plotted against the effective stop value of the 
fens, ta, corrected to a distance of 4 ft. 



graphic test were made of a large source at a considerable distance 
from the lens, a plot of negative- density versus //value should give 
only one curve regardless of the lens used, which would be the dy- 
namic sensitometric gamma characteristic of the film used. It was 
more convenient, however, to photograph a smaller object at a dis- 
tance of only a few feet from the camera, which introduces a small 



354 C. R. DAILY Vol 46, No. 5 

TABLE 3 

Focal Length Data on a Number of Lenses, Distant Object Photographic Method 
Focal Length 



Lens 


Nominal 
(Mm) 


Measured 
(Mm) (In.) 


Tan 0/2 


A/2 In. 


A 


25 


25.6 


1.01 


0.356 


0.359 


B 


35 


35.5 


1.40 


0.259 


0.378 


C 


50 


48.9 


1.92 


0.364 


0.364 


D 


75 


74.8 


2.95 


0.138 


0.410 


E 


100 


102.0 


4.01 


0.0945 


0,379 



correction factor because of the variation in magnification. This 
correction can be expressed as 



where t m is the //value of the lens at infinity focus, in the electrically 
calibrated value, t d is the //value at a distance d, d is the distance from 
lens to screen, F is the focal length of the lens, and n is the distance 
from lens to screen, expressed in focal lengths, i.e., d/F. For the 
4-ft screen distance used, this correction factor is listed in Table 4. 

TABLE 4 
F (In.) n U 

1 48 1.02 

2 24 1.04 

3 16 1.06 

4 12 1.09 

Fig. 9 shows a plot of the electrically obtained //values of several 
lenses, corrected to 4 ft, versus the density of the film as measured in 
the center of the film area. The agreement of calibrated //numbers 
and photographic densities is seen to be quite close. These tests 
have also been repeated at 10-ft distances to the screen with a similar 
close agreement. Fig. 10 shows a plot for two of the lenses used in 
Fig. 9, where the density of the film is plotted in 2 ways : "(a) the // 
numbers as originally marked on the lenses, corrected, of course, by 
Eq. (3) ; and (b) plotted against t d in the same manner as in Fig. 9. 
This plot illustrates that lens C as marked would give quite erroneous 
exposure values. 

Proposed Improvements. A number of improvements are cur- 
rently planned for this measuring apparatus to provide improved 
convenience of operation and to provide new information. The re- 



May, 1946 



A LENS CALIBRATING SYSTEM 



355 



calibration of lenses will be expedited by the use of a special attenua- 
tor whose loss per step will be the electrical equivalent of l / stop. 

With such an attenuator, the operators can start with a reference 
deflection on the volume indicator corresponding to a predetermined 

























x' 


1 7 




















/ 


e 






. 
















/ 




15 


















i 







13 
















/ 


/ 

e 






%_ 1 1 















/ 










i 

i i 9 






( 


y 


/ 















^ a 






/ 










CALIBRATED (-^ ) 







Q / 












LENS "A" 


z 




/ 












LENS "C" 




X 


/** 


( 










AS MARKED (f) 




x 














LENS "A" - 
LENS "C" - 


5 



































































20 15 10 9 8 7 6 5 4 3 2 I 

tar VALUES 

FIG. 10. Comparison of photographic data for 2 
lenses, where the originally marked and electrically cali- 
brated stop values are plotted against the density in the 
center of the field. 



maximum diaphragm opening on the lens and then start changing the 
attenuator by quarter- or half-stop intervals, setting the diaphragm 
to re-establish reference deflection again and scribe the lens for the 
desired marking. 

The fall-off of image brightness for off-axis rays will be checked on 
this apparatus when provision is completed for pivoting the inte- 



356 C. R. DAILY 

grating sphere around the center of the front face of the lens under 
test. 

The equipment is also being modified to mount 2 phototubes on the 
sphere with filter attachments for each cell. This change will permit 
adjustment of sensitivity and color balance of the system to make it 
substantially flat over the useful color range. Selective filters can 
then be inserted in the light path ahead of the lens to permit the pre- 
cise calibration of the transmission of a lens with respect to color, in- 
cluding the infrared region. Spectrophotometric measurement has 
shown that coated lenses may vary appreciably in their transmission 
characteristics over the useful spectrum. The sensitivity of the equip- 
ment may be high enough so that the transmission of a lens can be 
calibrated directly from a monochromatic light source. The narrow 
band-pass filter used in the electrical system rejects practically all 
noise except in the narrow band being measured, permitting a sub- 
stantial increase in gain of the amplifier system used. 

Summary. A description has been given of a new form of lens 
transmission calibrating equipment which should give reproducible 
results in any laboratory. The effective //values are obtained in 
terms of a known stop diameter. It may be desirable to continue to 
use the small letter /for the diaphragm marking of a lens based on the 
geometry of that lens and introduce a new symbol such as / to repre- 
sent the equivalent calibration in terms of the transmission of light. 

The writer wishes to express his appreciation to Lars Moen for the 
many suggestions and help in carrying out the preliminary part of this 
investigation. 

REFERENCES 

1 CLARK, D. B., and LAUBE, G.: "Twentieth Century Camera and Accessories," 
/. Soc. Mot. Pict. Eng., XXXIV, 1 (Jan., 1941), p. 50. 

2 BERLANT, E.: "A System of Lens Stop Calibration by Transmission," J. 
Soc. Mot. Pict. Eng., 46, 1 (Jan., 1946), p. 17. 

3 SILVERTOOTH, E. W. : "Stop Calibration of Photographic Objectives," /. 
Soc. Mot. Pict. Eng., XXXIX, 2 (Aug., 1942), p. 119. 



ANSCO COLOR FOR PROFESSIONAL MOTION PICTURES 
H. H. DUERR AND H. C. HARSH** 



Summary. The 3 new Ansco Color Films which are designed for producing full 
color motion picture release prints are described. These films are (1} Ansco Color 
Type 735 (Camera Film), (2) Ansco Color Type 132 (Duplicating Film), and 
(3) Ansco Color Type 732 (Release Film)^ Methods for making second genera- 
tion dupes, special effects, lap dissolves, etc., are discussed, and the procedure for 
printing sound tracks is outlined. 



The basic principles of the Ansco Color process have been described 
previously. We will, therefore, limit the discussion of these principles 
to a brief review of the fundamentals of the process so far as they are 
necessary for the proper understanding of the application of Ansco 
Color to motion picture production. 

The Ansco Color process is an integral sub tractive color process using 
the method of dye coupling for the production of dye images in a 
multilayer material. Colorless color-forming components are incor- 
porated in the emulsion layers. It is the unique and very important 
property of the color-formers in the Ansco Color process that they are 
of a molecular structure which renders them nondiffusing. The color- 
formers are immobilized in their respective emulsion layers and do 
not bleed into adjoining layers. 

The layer arrangement of Ansco Color Reversible Film is shown in 
Fig. la and Ib. The film base, which can be either cellulose nitrate 
or acetate, carries an antihalation layer, followed by the red-sensitive 
emulsion layer. This emulsion layer also contains a colorless dye- 
forming component which, upon development in a suitable color de- 
veloper, develops an image in color, complementary to the color sen- 
sitivity of the layer. In the case of the red-sensitive emulsion layer, 
the color is blue-green or cyan. For reasons of simplicity, this layer 
is usually referred to as the "cyan" layer. 



* Presented Oct. 17, 1945, at the Technical Conference in New York. 
"* Ansco, Binghamton, N. Y. 



357 



358 



H. H. DUERR AND H. C. HARSH 



Vol 46, No. 5 



The green-sensitive middle layer contains a color-former which, 
upon development, produces a magenta image, therefore called 
"magenta" layer. A yellow filter layer, coated on top of the magenta 
layer, absorbs all blue light, which would normally affect also the 
cyan and magenta layer, and therefore has to be filtered out in order 
to obtain the desired separation of color in these layers. The top emul- 
sion layer is blue-sensitive only and the nondiffusing color-former in 
this layer develops to a yellow image. This layer will be referred to as 
the "yellow" layer. 

The dye formers or color-formers in all 3 layers have been carefully 
selected so that they develop to a cyan, magenta, and yellow color, re- 



BLUE SENSITIVE LAYER 

YELLOW FILTER L 
GREEN SENSITIVE LAYER 
RED SENSITJVE LAYER - 
ANTI HALO LAYER 
BASE- 




FIG. 1. Layer arrangement of Ansco Color Film. 

spectively, in one color developing step. This greatly simplifies the 
processing of Ansco Color Film and makes it possible to have the 
complete processing done by the consumer with developing equip- 
ment which is very similar to that regularly used for black-and-white 
reversible development. 

The fundamental principles of the Ansco Color process are appli- 
cable to a great variety of color products. Ansco Color Reversible 
Film for daylight and tungsten light has been in use for some time in 
the form of 16-mm and sheet film. These materials are being manufac- 
tured for use primarily for direct projection and, therefore, have 
gradation characteristics which make them particularly well suited 
for this purpose. These film types, however, are not very satisfactory 
for the motion picture industry, where the requirements are essentially 
different. The most important requirement for a color transparency 



May, 1946 



ANSCO COLOR FOR MOTION PICTURES 



359 



suitable for 35-mm motion pictures is that it lends itself to the print- 
ing of first and second generation duplicates with a minimum loss in 
color brilliance and fidelity. 

These considerations call for a camera film which is quite different 
in gradation, color balance and other characteristics from the regular 
Ansco Color Film. Ansco Color Type 735 is the new film material 
designed and developed to meet these specific requirements of the 
motion picture industry. 

Fundamentally, Ansco Color Type 735 is quite similar to the regu- 
lar Ansco Color Film and the layer arrangement is the same as shown 






2A 
2.0 





\ 

\ > 


















\ 


\\ 






AN. 


)co COLOR FIL/*\ 
rvpe-235 

Co COLOR 
TvPc-735 

AVEPA F|LA\ 


. 






\ 


\ 




ANS 










\ 

\ 


\ 


Cf 












\\ 

\ 


\ 
















\ 


s 


\ 
















\ 


^ 


\ 


















-..^X 





FIG. 2. Characteristic curves of Ansco Color Camera Films. 

in Fig. la. It differs from this film primarily in that the gradation 
is considerably softer, the grain is finer, and the color balance is 
purposely slightly off -neutral. Fig. 2 shows a comparison of the H and 
D curves of Ansco Color Type 735 and the regular Ansco Color 
Daylight Film Type 235. This new material is designed to provide 
a film for exposure in the camera which is ideally suited for making 
release prints on Ansco Color Release Film Type 732. Ansco Color 
Type 7.V5 is not intended for projection and its use in motion picture 
practice should parallel the use of the original negative in black- 
and-white motion pictures. Because it is a positive color transparency, 
it is, however, possible to judge immediately after development the 
color rendition and other pictorial effects of the scene. 



360 



H. H. DUERR AND H. C. HARSH 



Vol 46, No. 5 



Ansco Color Camera Film is available on both nitrate and acetate 
base, and designated 735 and 835, respectively. 

Ansco Color Camera Film is balanced for exposure by daylight 
and the best color rendition on exterior exposures will be obtained 
in bright sunlight. For studio exposures, excellent results are ob- 
tained with key-light provided by high-intensity carbon arcs which 
are modified by Y-l gelatin filters and fill-light by tungsten lamps 
for color photography filtered with Macbeth Whiterlite filters. The 
spectrogram shown in Fig. 3 gives the relative response of the film 
to the visible region of a daylight spectrum. 

Suggested meter settings for the exposure of Ansco Color Type 735 
are Weston 8, or G. E. 12. 

For optimum print results, it is desirable that the Ansco Color 
original be slightly underexposed or somewhat heavier in density than 




40 44 48 52 56 60 64 68 

FIG. 3. Wedge spectrogram of Ansco Color Camera Film, Type 735. 

is the usual practice when exposing a transparency for screen projec- 
tion. The reason for preferring a heavy original is to maintain as 
much of the exposure as possible on the straight-line portion of the H 
and D curve and avoid inaccurate color reproduction which can result 
from exposures which fall predominantly in the toe region. 

The processing of Ansco Color Type 735 is almost identical to 
that which has previously been described by Forrest. 1 The only 
variation is a somewhat shorter developing time in both the first and 
the color developer. In this connection we believe it is indicative of 
the simplicity of the Ansco Color process that several laboratories have 
converted existing black-and-white machines for the processing of 
Ansco Color Film and in all instances very little difficulty has been 
encountered. At least in one case, the very first roll of Ansco Color 
Film which was developed on a converted black-and-white machine 
was of excellent quality. 

(1) Direct Prints from Original Color Transparency. As pre- 
viously mentioned, the release printing stock for an Ansco Color 



May, 1946 



ANSCO COLOR FOR MOTION PICTURES 



361 



Type 735 original is Ansco Color Release Film Type 732. This film 
is also of the reversible type and while fundamentally similar to the 
other Ansco Color reversible films, it is characterized by a relatively low 
speed, very fine grain and special sensitization for printing. This 
printing stock can be developed to a high maximum density to ob- 
tain optimum color brilliance. Fig. 4 shows a typical H and D curve 
for this film. Fig. 5 shows a wedge spectrogram of the Ansco Color 
Release Film. There are relatively sharp sensitivity peaks in the green 
and red regions and a partial gap between these peaks. Good separa- 



/ANSCO COLOR 
RELEASE RLAA 
TYPE - 732 



FIG. 4. Characteristic curve of Ansco Color Release Film. 



tion of the peaks of sensitivity is very essential in a printing film in 
order to obtain faithful color reproduction. 

Ansco Color Release Film will be available on both nitrate and 
acetate base and designated 732 and 832, respectively. 

Most motion picture printers which are suitable for printing 
present-day black-and-white positive stocks can be readily adapted 
to print Ansco Color Release stock. If not already available, the 
following features should be provided on a printer to make it suitable : 

(1) A light source which operates at a color temperature of approximately 
3000 K. 

(2) A means for inserting printing filters into the light path quickly and con- 
veniently. 



362 



H. H. DUERR AND H. C. HARSH 



Vol 46, No. 5 



(5) A condenser lens system for the light source in order to concentrate the 
light at the aperture. Ansco Color Release Film, with the printing niters in 
place, will require 2 to 4 times the light needed for printing black-and-white posi- 
tive fine-grain stock. 

(4) It is good practice to provide an air blast or fan as a means of dissipating 
the heat from the lamp house in order to avoid damage to the filters and film. 

Using a regular black-and-white printer with the modifications just 
described, the printing of the Ansco Color original onto Ansco 
Color Release Film requires the insertion of filters to balance the 
color quality ,of the light source. A standard series of Ansco color 
compensating niters, in varying densities of yellow, magenta, and 
cyan, are available for this purpose. Considerable control of the 
color balance of the release print is possible by the selection of these 
printing filters. 




36 40 44 46 52 56 6O 64- . 6B 72 

FIG. 5. Wedge spectrogram of Ansco Color, Type 732 Release Film. 



The processing of Ansco Color Release Film is carried out on the 
same developing machine and in the same solutions as used for the 
Ansco Color original. Adjustments of the developing times to 
suit the particular machine conditions are necessary. 

So far we have discussed the camera film and the release printing 
film for the Ansco Color process without referring to methods for 
including optical lap dissolves, wipes, and special effects where second 
generation duplicates of the original Ansco Color will be involved. 

(2) Prints from Duplicates. It is generally recognized that in 
color reproduction each printing step results in a noticeable deg- 
radation in color. For this reason it is desirable to reduce the 
number of printing operations in color photography to a minimum. 
However, in motion picture practice it is not feasible in many instances 
to print from the original color transparency. This is particularly 
true in those cases where special effects, such as lap dissolves and 
wipes, have to be incorporated in the sequence of the picture, also for 
foreign releases where 'it is essential that a master dupe is available 



May, 1946 



ANSCO COLOR FOR MOTION PICTURES 



363 



for release printing. Two methods for making master dupes have 
been worked out for this purpose. 

The first method consists of straightforward optical printing of the 
Ansco Color original onto Ansco Color Type 132 Duplicating Film. 

The duplicating stock Type 132 requires about the same exposures 
as the release stock 732, that is, approximately 2 to 4 times the light 
needed for regular positive fine-grain stock. The film is processed in 
the same solutions as the Type 735 original. The developing time in 
the first and second developer is shorter. A duplicate is obtained 



2.8 



ANSCO COLOR 
DUPLICATING (1 32) 



FIG. 6. Characteristic curve of Ansco Color Duplicating Film. 



which is substantially equal in contrast to the camera original. 
This first generation duplicate can then be interspliced with the 
original and used for release printing on Ansco Color Type 732 
Release Film. The H and D curve of the Type 132 Duplicating 
Film is shown in Fig. 6. 

The fact that the original as well as the dupe and the release print 
stock can all be developed in the same machine and in the same solu- 
tions represents a very essential simplification of the Ansco Color 
process. As pointed out before, there are differences in the develop- 
ing times for these 3 color films, and in order to allow a more exact 
comparison, the approximate developing times are listed : 



364 



H. H. DUERR.AND H. C. HARSH 



Vol 46, No. 5 



Ansco Color Type 755 or 835 (Camera Film) 
Ansco Color Type 132 or 232 (Duplicating Film) 
Ansco Color Type 732 or 832 (Release Film) 



First 
Developer 

9 min 

8 min 

10 min 



Color 
Developer 

11 min 

9 min 

18 min 



The developing times shown are only approximate, since the exact 
time depends very largely upon the machine speed and the solution 
agitation in the machine. 

There will be an inevitable loss in color brilliance in the second 
generation duplicate prepared by this method. However, the loss is 



48 
2A 
& 
1.6 
12 
.8 
.4 

o 
FIG. 7. 






















PANCHROMATIC 














MASKING FILM 
TYPE -154 
GAAW* .40 




















































^ 


,> 


" 










^ 


-"" 













-- 


""" 












Characteristic curve of Ansco Masking Film, Type j 



probably not serious enough to preclude its use for certain lap dis- 
solves, wipes, and other special effects, especially if the subject of 
these special effects is of such a nature that a very critical judgment 
of color rendition is not possible. Because of the loss of color brilliance 
by this method, it is not recommended for making full-length master 
dupes. For this purpose and where good color reproduction is de- 
sired on special effects, the following second method is preferred. 

(3) Prints from Masked Duplicates. In order to counteract 
the color degradation, the second method of making a master dupe 
employs a black-and-white silver mask. It is not the purpose of this 
paper to go into the details of the theoretical requirements of masking, 



May, 1946 



ANSCO COLOR FOR MOTION PICTURES 



365 



but rather restrict this discussion to the recommendation of a simple 
procedure for masking Ansco Color Type 735. 

A special low-shrink, panchromatic, black-and-white film has been 
developed for masking in connection with the Ansco Color process. 
The characteristics of this material are such that the required mask- 
ing densities are obtained with the least amount of critical control. 
For this reason, the gamma infinity of the material is adjusted to the 



OR/GINAU 



YNCHRONIZED REGISTER PiNS 




LIGHT SOURCE 



PROJECTION HEAD CAMERA HEAP 

FIG. 8. Schematic view of optical printer. 



masking requirements. In Fig. 7, the characteristic H and D curve 
of this special masking film is shown. In order to insure good regis- 
tration, the same printing equipment should be used for the print- 
ing of this color correction mask which later on is used for the print- 
ing of the masked master dupe. In Fig. 8, a schematic outline of 
the type of optical printing equipment which can be used for this 
purpose is shown. The essential features of a suitable printer are 
2 synchronized intermittent movements with register pins which are 
combined with the necessary optical equipment. 



366 H. H. DUERR AND H. C. HARSH Vol 46, No. 5 

In making the black-and-white mask, the original and the mask are 
run in contact in the camera head while using the projection head 
empty as a light source only. A yellow filter is placed into the light 
path while printing the masking film. After the mask has been de- 
veloped in a regular negative developer, the master dupe is printed. 
In this operation, the original is run in the projection head and opti- 
cally registered with the black-and-white silver mask which is now 
run in the camera head in contact with the Ansco Color Type 132 
Duplicating Film. 

The printer must be equipped with a viewer or other suitable means 
so that the registration of the original and the mask can be checked 
before printing the master dupe. The conformed master dupe can be 
made in one printing operation, even though special effects may be 
necessary. In the case where special effects that require mattes are 
to be inserted, these should be run in the projection head with the 
original. This method will yield a conformed master dupe that will 
show little or no loss in color brilliance. 

This masked master dupe is then used for printing of the release 
prints, using Ansco Color Type 732 Release Film in a regular con- 
tinuous printer with provisions for the insertion of filters and a 
stronger light source, as described earlier. 

So far, only the pictorial part of the process has been discussed, 
but it is realized that methods of obtaining good sound are of equal 
importance. Since the release printing stock is a reversible film, a 
positive black-and-white track is required for printing. The ideal 
way to obtain the black-and-white positive would be a direct-positive 
recording. However, equipment to record to a direct positive is not 
generally available, and the following method has been 'found almost 
equally satisfactory and is the one recommended. The recording 
head of the sound equipment is moved so that the negative recording 
is obtained on the opposite side of the film. This negative is then 
printed onto black-and-white positive stock, which will then have the 
sound track in the proper position for printing directly onto the Ansco 
Color Type 732 Release Film in the conventional manner. 

Dye tracks, especially those obtained by the dye coupling method, 
have a relatively low absorption in the infrared region. Therefore, 
the conventional infrared-sensitive photocell, for example type 868, 
is not too well suited for these dye tracks and a loss in volume amount- 
ing to approximately 6 db is encountered. This loss in volume, while 
serious, still comes within the" range where adjustment can be made 



May', 1946 ANSCO COLOR FOR MOTION PICTURES 367 

by fader setting on most 35-mm projection equipment. Fortunately, 
within the last few years the development of blue-sensitive photocells 
has progressed and cells are available today which are ideally suited 
for dye tracks and will play normal silver tracks with approximately 
the same volume so that interchange of tubes is not required. This 
photocell, which is at present available from the Radio Corporation 
of America, is designated as the IP -37. 

Summarizing, the Ansco Color process is capable of producing full 
color motion picture release prints, including commonly used effects, 
with only minor changes in equipment which is now used extensively 
for black-and-white motion pictures. We believe that the Ansco 
Color process offers a relatively simple method for making motion 
pictures in color which can be readily mastered by those skilled in black- 
and-white motion picture techniques. 

REFERENCE 

1 FORREST, J. L.: "The Machine Processing of 16-Mm Ansco Color Film," /. 
Soc. Mot. Pict. Eng., 45, 5 (Nov., 1945), p. 313. 



SENSITOMETRIC EVALUATION OF REVERSIBLE COLOR 

FILM* 



RONALD H. BINGHAM** 



Summary. Two considerations are involved in the sensitometric evaluation of 
reversible color film; we are interested in determining the gray-scale characteristics of 
the material, as well as its faithfulness of color rendition. Equations have been de- 
rived relating the equivalent densities of the 3 layers to the color den sitometer readings. 
The 3 curves showing equivalent densities as functions of log exposure indicate speeds, 
contrasts, maximum densities, and color balance of the film. Color rendition problems 
are investigated using the principles of colorimetry. 

The equivalent densities of a reproduction depend upon the spectral sensitivities of 
the material as well as the spectral characteristics of the light source and color sample; 
the color of the reproduction, however, depends not only upon the equivalent densities 
but also upon the spectral characteristics of the dyes and of the projection source. A 
typical problem is that of the effect of change of contra^ upon the color renditions. 
Unit contrast, although giving lighter reproductions, involves saturation errors which 
are prohibitive. A contrast of 1.5 is found to more nearly balance the errors of light- 
ness, hue and saturation. 

Introduction. In the sensitometric evaluation of color film we 
seek the answer to 2 questions: first, how well does the material 
reproduce a scale of grays, and second, how well does it render the 
whole gamut of photographically important colors ? ' The first of 
these questions concerns certain sensitometric characteristics of the 
individual layers, their speeds, contrasts, and toe and shoulder char- 
acteristics, while the second is related, in addition, to the spectral 
aspects including the spectral sensitivities of the layers and the spec- 
tral densities of the primary dyes. 

Gray-Scale Sensitometry. With color film, just as with black- 
and-white film, gray-scale sensitometry centers about the char- 
acteristic curve which expresses the relationship between the den- 
sity of a given layer and the logarithm of the exposure. In the 
case of subtractive color film we have 3 such curves (Fig. la and Ib) 
which are related to the yellow, magenta, and cyan layers, respec- 



* Submitted Jan. 8, 1946. 
** Research Laboratories, Ansco, Binghamton, N. Y. 



368 



EVALUATION OF REVERSIBLE COLOR FILM 



369 



lively. They are plotted from measurements made on a sensitometric 
strip which has been exposed and developed under standardized 
conditions. 

In exposing the sensitometric strip, we require an intensity scale 
sensitometer which has a light source of the correct spectral energy 
distribution and which yields an exposure of the order of x /25 sec. 
For tungsten-type film, the incandescent light source need only have 
the required color temperature of 3200 K, while daylight-type film 



10 



1.0 



% 



V 

n 

V 



3.0\ 



" 



\ 



\ 

H 



-a.o -1.0 o 

LOG. EXPOSURE 

EQUIVALENT DENSITY CURVES 
OF A BALANCED STRIP 



LOG. EXPOSURE 

EQUIVALENT DENSITY CURVES 
OF AN UNBALANCED STRIP 



(a) (b) 

FIG. 1. Equivalent density curves of balanced and unbalanced sensitometric 
strips yellow ; magenta x x x ; cyan o o o. 



requires a lamp-filter combination having, as nearly as possible, the 
same energy distribution as daylight. 

As a means of expressing quantitatively the absorption owing to 
the dye in each of the 3 layers, Heymer and Sundhoff 1 and Evans 2 
developed the concept of equivalent density, defining it essentially as 
follows: "In a sub tractive color process, the equivalent density 
of the primary dye in a layer is equal to the gray density which would 
be obtained if it were superimposed upon the required amounts of 
the other two primaries to form a visually gray combination." The 
usefulness of^this concept becomes increasingly clear when we note 
that, by definition, a perfectly balanced strip one having all steps 



370 



R. H. BlNGHAM 



Vol 46, No. 5 



gray would have 3 curves which would coincide as in Fig. la. On 
the other hand, with an unbalanced strip (Fig. Ib), any step which 
departs from gray results in inequality of the 3 equivalent densities, 
a reddish step, for instance, having a deficiency of cyan, etc. 

In determining the 3 characteristic curves we are unable to separate 
the layers and hence are forced to make our measurements on all 3 
superimposed layers of the sensitometric strip. Evans 2 has described 



UNIT EQUIVALENT 
DENSfTIES 

.Co) 




400 



500 



WAVELENGTH 




EQUIVALENT DENSITIES 
J,m ANDc 

(b) 



400 



500 

WAVELENGTH 



FIG. 2. Spectral density curves of 3 primary dyes. 

a visual type of color densitometer patterned after the Capstaff- 
Purdy instrument which uses wedges made of the primary dyes of the 
process. Another approach which has proved of considerable value 
involves the measurement of the integral spectral densities* at 3 
wavelengths. For this purpose, accurate work such as determina- 
tion of calibration constants, is carried out on a spectrophotometer. 
In our development and control work, however, we have preferred 
to use a photoelectric instrument such as the one which was de- 



* The word "integral" as used here implies that the spectral density measure- 
ment is made on the superimposed yellow, magenta, and cyan layers. 



May, 1946 EVALUATION OF REVERSIBLE COLOR FlLM 371 

scribed by Sweet. 3 This instrument employs a photomultiplier tube 
circuit in combination with an incandescent source and 3 narrow 
band-pass optical filters. With it, 3 spectral density readings are 
made using nearly monochromatic light, with each of the wavelengths 
being located near the peak of one of the 3 primary dyes.* 

Typical primary dye curves as shown in Fig. 2a evidence a con- 
siderable amount of overlap so that, for example, a reading made at 
430 m/z is influenced not only by the amount of dye in the yellow 
layer but also, to a considerable extent, by the amounts in the other 
2 layers as well. The curves of Fig. 2a have been drawn for dyes 
each of which have an equivalent density of 1.0. Thus, by definition, 
the corresponding sample must be gray and have a visual density of 
1.0. In preparation for the derivation to follow, let us consider the 
spectral densities of each of the 3 layers at the peak wavelengths. 
For the yellow layer these are designated** J b , J ff , and J r ; for the 
magenta, M b , M g , and M r , while the corresponding cyan densities are 
C?,, C g , and C r . 

Now let us consider another sample, whose curves are shown in 
Fig. 2b, having unknown equivalent densities, j, m, and c, which are 
to be determined. It has been shown experimentally that Beer's 
law holds sufficiently close for the typical primary dyes under con- 
sideration, hence the spectral densities of the primary dye curves 
of Fig. 2b-are proportional to those of Fig. 2a. Furthermore, ex- 
perimental work has shown that the 3 proportionality constants are 
just the equivalent densities of the layers in question, so that, for 
example, the spectral densities of the yellow layer at the peak wave- 
lengths are jj b , jj g , and j J T , while the corresponding densities of the 
other 2 layers are mM b , mM g , mM r , cC b , etc. As a means of evaluating 
the equivalent densities of such a 3-layer sample we read its integral 
spectral densities B, G, and R at the peak wavelengths. Each of these 
values equals the sum of the corresponding spectral densities of the 
individual layers. Using the notation developed above, we have 



B = jJ b 

G = jj a + mM g + cC , 

R = jJ r + mM r + cC T . 



* These 3 wavelengths will be referred to as the "peak" wavelengths; with the 
instrument referred to above they are 430, 540 and 660 mju. 

** The letters / and j are used instead of Y and y to represent densities in the 
yellow layer in order to avoid conflict with the notation of additive colorimetry. 



372 R. H. BINGHAM Vol 46, No. 5 

These equations involve the 3 unknown equivalent densities j, 
m, and c in terms of the measurable spectral densities, B, G, and R as 
well as the constants J b , J g , etc., which are determined by the curves 
of Fig. 2a. In order to determine the equivalent densities we simply 
solve for ^', m, and c, obtaining 

j = kiB k z G k 3 R, 

m = -k t B + k,G - ktR, (2} 

t = -k 7 B - k s G + kgR. 

Eqs (2) have been written with the proper signs so that the constants, 
ki t ky-kv will be positive. 

The constants of- Eqs (2) are determined from measurements on 
representative samples of the color film. The spectral density curve 
of each of the dyes is determined by measuring samples in which 2 lay- 
ers have been strongly exposed to leave only one dye. Frdm these 
data, the curves corresponding to unit equivalent densities are calcu- 
lated using the standard equations of colorimetry. 4 ' 5 Having deter- 
mined the constants, we are ready to use Eqs (2) as a basis for routine 
sensitometry. The sensitometric strip is read at each step to obtain 
the densitometer densities B, G, and R. These data are substituted 
into Eqs (2) yielding the corresponding equivalent densities of each 
step. In routine application of the equations, charts are used which 
permit rapid calculation of the equivalent densities from the values 
of B, G, and R. 

The equivalent density curves of Fig. Ib may now be interpreted 
in the light of the above discussion. Considering first the curves 
representative of a perfectly balanced material, we note that we are 
able to carry over directly the concepts of black-and-white sensi- 
tometry such as speed, contrast, maximum density, toe and shoulder 
characteristics, etc. We thus see that equivalent density is merely 
a generalization of the usual definition of density. 

Turning our attention to the curves of the unbalanced material, 
we note certain parts of the log exposure range where one curve is 
higher than another. Thus, in the shoulder region we have the yellow 
falling above the other two. This may be interpreted to indicate 
that the shoulder will be yellowish. Similarly, we find a deficiency 
of cyan in the intermediate densities with the result that the repro- 
ductions of middle tones would be slightly brownish. Since in this 
case the contrasts of all 3 layers are about equal, the reproductions 
would have been considerably improved had the cyan layer been a 



May, 1946 EVALUATION OF REVERSIBLE COLOR FlLM 373 

little slower or better still, of course, if the other two had been faster. 

Determination of Color Rendition. Having considered the gray- 
scale aspects of the problem, let us turn our attention to the prob- 
lem of color rendition. It is well known that the proper rendition 
of a gray-scale is by no means the only requirement which must be 
fulfilled by a color reproduction material, since in the worse cases, for 
example, sensitizing dyes could be used which would yield completely 
false colored pictures. It should be pointed out that theoretical as 
well as experimental work has shown that the best we can do is to 
choose the film characteristics in such a way as to give a minimum of 
distortion, since it can be shown that perfect color reproduction by a 
reversible material is impossible without negative sensitivities over 
certain parts of the visible range of wavelengths. 

Experimentally the process of evaluating a film as to its faithful- 
ness of color reproduction is straightforward. We make up a color 
chart from a series of reflecting color samples whose colors have 
previously been calculated from their spectral reflectance curves. 
This color chart is illuminated by a light source of carefully selected 
and controlled spectral energy distribution, and photographed, yield- 
ing a series of reproductions which are measured with a spectro- 
photometer. The reproduction colors are calculated from the result- 
ing spectral transmittance curves by standard colorimetric meth- 
ods. 4 - 5 Detailed comparisons between samples and reproductions 
may be made and the over-all color rendition may be determined by 
the methods outlined later in this paper. 

A color rendition problem of considerable importance is that of 
determining the difference in color reproduction to be expected 
if we should either change the photographic material in the course 
of manufacture or change the processing procedure. In many cases, 
it becomes desirable to determine the reproduction errors by theo- 
retical methods. In order to illustrate the general approach let us 
consider the problem of relative color distortion at 2 different con- 
trasts, 7 = 1.0 and 7 = 1.5. In other words, the question which we 
raise is, "Which of these 2 contrasts will give the better color repro- 
duction?" This example has been selected for theoretical discussion 
because the conclusions have been checked experimentally. 

Without going into the mathematical details it may simply be 
said that we start with the spectral sensitivity curves which have 
been experimentally determined. In addition, we must know the 
energy distribution curve of the illuminant as well as the spectral 



374 R'. H. BlNGHAM Vol 46, No. 5 

densities of the primary dyes. We adopt a series of reflection color 
samples whose reproductions are to be calculated, determining first 
the exposures to be expected in the 3 layers and from them the equiv- 
alent densities. For the straight-line portion of the characteristic 
curve, the density is proportional to 7, thus, 

j = y log E 6 , 
m = 7 log Eg, (3) 

C = 7 log E T , 

from which we see that over this exposure range the effect of the 
change of contrast is just to multiply all equivalent densities by 1.5. 

In these equations the exposures in the blue, green, and red sensitive 
layers, E b , E g , and E r , are expressed in such units that for white, E b = 
E = E r = 1.0. Thus in Eqs (3), log & = logE, = log r = 0. 

The equivalent densities which were calculated for the reproduc- 
tions of 5 saturated Munsell colors are shown in Table 1. 

TABLE 1 

Equivalent Densities of the Reproductions of 5 Munsell Samples at 7 = 1.0 and 

7 = 1.5 
Sample Equivalent Density 





Munsell 




7 = 1.0 






7 = 1.5 




Color 


Notation 


j 


m 


c 


j 


m 


c 


White 


N 10 




















Red 


5R 4/14 


1.31 


1.37 


0.27 


1.97 


2.05 


0.41 


Yellow 


5 Y 8/12 


1.30 


0.20 


0.18 


1.95 


0.29 


0.27 


Green 


5G5/8 


1.05 


0.64 


1.12 


1.58 


0.97 


1.68 


Blue 


5B4/8 


0.67 


0.95 


1.28 


1.00 


1.42 


1.93 



Purple 5P4/12 0.47 1.10 0.55 0.70 1.65 0.82 

Having obtained the reproductions in terms of the amounts of 
the yellow, magenta, and cyan primary dyes, we are next interested 
in determining the actual color specifications and finally the color 
differences. In order to carry out the necessary colorimetric inte- 
grations we need an expression for the spectral transmittance of a 
sample of equivalent densities, j, m, and c. From the definition of 
equivalent density, we have for the spectral density curve of the 3- 

layer sample 

D = j/i + mMi + cC lt (4) 



where Ji, Mi, and C\ are the spectral density curves for unit equiva- 
lent densities. It then follows from the definition of density that the 
spectral transmittance is given by 

T = 10- = iQ-Wi + m 



May, 1946 EVALUATION OF REVERSIBLE COLOR FlLM 375 

Using the primary dyes of Fig. 2a, the tristimulus specifications were 
calculated by the usual integration formulas 4 ' 5 yielding the data of 
Table 2 in which are given the luminous reflectance Y and the tri- 
chromatic coefficients x and y for each of the 5 samples as well as the 
corresponding specifications of their reproductions. 

TABLE 2 

Tristimulus Specifications of 5 Munsell Samples and Their Reproductions at y 1.0 

and 7 = 1.5* 

Y x y 

White Sample 1.000 0.322 0.338 

NW 7 = 1.0 1.000 0.322 0.338 

7 = 1.5 1.000 0.322 0.338 

Red Sample 0.131 0.572 0.314 

5R4/14 7 = 1.0 0.117 0.499 0.337 

7 = 1.5 0.048 0.578 0.329 

Yellow Sample 0.570 0.466 0.488 

5Y8/12 7 = 1.0 0.536 0.426 0.451 

7 = 1.5 0.397 0.458 0.471 

Green Sample 0.190 0.258 0.431 

5G5/8 7 = 1.0 0.153 0.298 0.356 

7 = 1.5 0.061 0.304 0.410 

Blue Sample 0.113 0.205 0.252 

5B4/8 7 = 1.0 0.098 0.257 0.282 

7 = 1.5 0.032 0.235 0.254 

Purple Sample 0.121 0.299 0.191 

5P4/12 7 = 1.0 0.140 0.331 0.263 

7 = 1.5 0.054 0.331 0.237 

* In these calculations the illuminant was taken to be daylight, 6000 K, as 
determined from data of Parry Moon. 3 

In Fig. 3 the trichromatic coefficients are presented graphically. 
On this color diagram, points which represent different colors are 
grouped about the center (white) point, according to hue, with those 
representing colors of greater saturation being located farther from 
the center. Thus, for example, the 2 reproductions of the red sample 
are seen to be of about the same hue but of considerable difference in 
saturation. 

Analyzing the data, we conclude that the principal effects of in- 
creasing the contrast is to darken the samples and make them more 
saturated. In order to express quantitatively the difference between 



376 



R. H. BlNGHAM 



Vol 46, No. 5 



each of the samples and its 2 reproductions, we have calculated the 
errors in lightness, hue and saturation as they would appear under 
projection conditions. The formulas which were employed are of the 
type developed by Judd 6 and Hunter 7 at the National Bureau^of 



0.8 



MUNSELL COLOR SAMPLE 

* REPRODUCTION, i -1.0 
x REPRODUCTION, / * 1.5 




O.I 



0.2 



0.5 



0.6 



0.7 



0.3 04 

VALUES OF x 

FIG. 3. Chromaticities of reproductions calculated at 2 different contrasts. 
(NOTE : Dotted lines indicate sample and reproduction points associated with 
each color.) 

Standards. The results are given in Table 3. Each error in light- 
ness, hue or saturation has been computed in terms of the number of 
color steps which would be distinguishable between the sample and 
its reproduction, while the total color error is taken in the usual way 
as the square root of the sum of the squares of the 3 separate errors. 
The average color error is computed as a criterion of the over-all repro- 
duction. 



May, 1946 EVALUATION OF REVERSIBLE COLOR FlLM 



377 



In Table 3 a positive lightness error indicates a reproduction lighter 
than the sample, a positive hue error indicates a hue shift in the 
counter-clockwise direction as seen in Fig. 3, while a positive satura- 
tion error indicates reproduction more saturated than the sample. 



TABLE 3 

Color Errors for 5 Munsell Samples Reproduced at y 

Errors 



1.0 and 7 = 1.5 



Sample 
White 


Contrast 
7 = 1.0 


Lightness 




Hue 



Saturation 



Color 




N 10 


7 = 


1 


.5 






















Red 


7 = 


1 


.0 


-0, 


,7 


4 


.2 


-24. 





24. 


4 


5R 4/14 


7 = 


1 


.5 


-4 


.8 


3 


.6 


0. 


8 


6. 


1 


Yellow 


7 = 


1 


.0 


-0 


,8 


-1 


.0 


-18. 


7 


18. 


8 


5 Y 8/12 


7 = 


1 


.5 


-4. 


2 


2 


.3 


- 5. 


4 


7. 


5 


Green 


7 = 


1 


.0 


-1 


.5 


1 


.6 


-14. 


5 


14. 


7 


5G5/8 


7 = 


1 


.5 


-6. 


3 


-5 


.3 


- 6. 


2 


10. 


3 


Blue 


7 = 


1 


.0 


-0. 


.8 





.8 


-10. 


5 


10. 


6 


5B4/S 


7 = 


1 


.5 


-5. 


2 


2.8 


- 4.0 


7. 


2 


Purple 


7 = 


1 


.0 





,9 


4 


.6 


-17. 


3 


17. 


9 


5P 4/12 


7 = 


1 


.5 


-3 


.8 


5 


.7 


-11. 





13. 







7 = 


1 


.0 


, Average 


Color Error 


= 17. 


5 




7 = 


1 


.5 


Average 


Color Error = 8. 


8 



Examination of the data discloses the fact that the reproductions 
at 7 = 1.0 are lighter but considerably desaturated as compared 
with those at 7 = 1.5. The loss of saturation at the lower contrast 
is so serious as to more than offset the gain in lightness with the result 
that the over-all reproduction at 7 = 1.0 is considerably less satis- 
factory than that at 7 = 1.5. 

The author wishes to acknowledge valuable advice and encourage- 
ment on the part of Dr. H. Hoerlin, under whose direction the work 
was carried out. 



REFERENCES 

1 HEYMER, G., AND SUNDHOFF, D.: "Uber de Messung der Gradation von Far- 
benfilman," Agfa Zentral-Labor, Veroff., 5 (1937), p. 62. 

2 EVANS, R. M.: "A Color Densitometer for Subtractive Processes," /. Soc. 
Mot. Pict. Eng., XXXI, 2 (Aug., 1938), p. 194. 

3 SWEET, M. H.: "Densitometry of Modern Reversible Color Film," J. Soc. 
Mot. Pict. Eng., 44, 6 (June, 1945), p. 419. 



378 R. H. BlNGHAM 

4 MACADAM, D. L.: "The Fundamentals of Color Measurement," /. Soc. Mot. 
Pict. Eng., XXXI, 4 (Oct., 1938), p. 343. 

6 O.S.A. Colorimetry Committee: "Quantitative Data and Methods for Color- 
imetry," /. Opt. Soc. Am., 34 (1944), p. 633. 

6 JUDD, D. B.: "Specification of Color Tolerances at the National Bureau of 
Standards," Am. J. Psychol., 52 (1939), p. 418. 

7 HUNTER, R. S.: "Photoelectric Tristimulus Colorimetry with Three Filters," 
/. Opt. Soc. Am., 32 (1942), p. 509. 

8 MOON, P.: "Proposed Standard Solar-Radiation Curves for Engineering Use," 
/. Frank. Inst., 230 (1940), p. 583. 



A PHOTOTUBE FOR DYE IMAGE SOUND TRACK' 
ALAN M. GLOVER AND ARNOLD R. MOORE** 



Summary. In view of the use of dye image sound track on new-type color film, 
a phototube has been developed with characteristics suitable for sound reproduction 
f rom these films as well as from ordinary silver tracks. It is a gas-filled, high-sensi 
tivity phototube for use in standard reproducing equipment. The maximum spectral 
response occurs in the blue blue-green region of the spectrum. Details of construction, 
sensitivity, characteristic curves, spectral response, frequency response, and life are 
presented. 

Color is a subject of considerable interest to the Society of Motion 
Picture Engineers. Color complicates the problem of the sound track 
engineer. As an introduction to the discussion of a new photosensi- 
tive surface which will be of considerable value for -use with dye image 
sound tracks, a brief review of the characteristics of the phototubes 
now in general use is in order. 

The advent of sound with motion pictures in the late 1920's was 
in large measure made possible by the introduction of the caesium- 
silver-oxygen photosurface. Considerable effort had been expended 
in the previous decade in an attempt to raise the hitherto limiting 
sensitivity of the photosurf aces then available and, as is so frequently 
the case in technical progress, it is difficult to assay the part which the 
photosurface itself played in the new sound industry. Concomitant 
with the new photosurface were improved amplifying tubes and steady 
progress in the quality of the film, but the industry would have pro- 
gressed slowly without the new photosurface. 

The most advantageous characteristic of the caesium-silver-oxygen 
photosurface was its high sensitivity in the near infrared, a region of 
the spectrum in which the major portion of the energy of the incandes- 
cent light source is concentrated. About three-quarters of the total 
sensitivity of this photosurface to light from an incandescent source 
lies in the infrared. This sensitivity to infrared was of great advantage 
as long as the sound track consisted of a developed silver image which 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Radio Corporation of America, RCA Victor Division, Lancaster, Pa. 

379 



380 



A. M. GLOVER AND AI R. MOORE 



Vol 46, No. 5 



may be exposed to densities as high as three in this region. With the 
advent of dye image tracks a possibility, it immediately became appar- 
ent that the marked transparency of such tracks in the near infrared 
to which the S-l surface is so sensitive would seriously limit the modu- 
lation obtainable. This is true whether the sound track is of the vari- 
able-area or variable-density type. Other authors have already 
touched on this subject in a previous issue of the JOURNAL. 1 



100 



80 






FOR EQUAL VALUES OF RADIANT 
FLUX AT ALL WAVELENGTHS 



RCA-IP37 



3500 



7500 



4500 5500 6500 

WAVELENGTH -ANGSTROMS 

FIG. 1. Spectral sensitivity of phototube having S-4 
response. 



In 1940 there appeared information on a new photosurface, the 
sensitivity of which is largely concentrated at the short wavelength 
portion of the visible spectrum. 2 Most of the phototubes which have 
employed this surface are of the high-vacuum type and this is prob- 
ably the reason for the slow acceptance of such tubes by the motion 
picture industry. The gas-filled phototube has been popular in the 
industry for 3 reasons : high sensitivity with resultant high signal-to- 
noise ratio, adaptability of the voltage sensitivity of the gas tube as a 



May, 1946 PHOTOTUBE FOR DYE-IMAGE SOUND TRACK 



381 



volume control, and the lower impedance level of the gas tube. Al- 
though the cause is not yet completely known, some difficulty has been 
encountered in introducing inert gas into phototubes containing the 
S-4 surface. Such tubes have suffered from short life. However, this 
paper will outline the characteristics of a new tube, the RCA 1P37, a 
gas-filled phototube of good life whose properties are such as to indi- 
cate the broad possibilities for its use with dye image sound tracks. 



Of) 




\ 1 

\s-4(ip: 


k7) 


< 


f 




^ 


\ 






! 




/ 










X 






1 


1 


^ 














| 


/ 














Q. 
(0 
Ld 

(T 
UJ 




u 
















1 














UJ 

cr 
20 






1 






























i 


/ i 














\ 


d^r 






S-l ft 


68) 

I 








X 



3000 5000 7000 9000 IK 

WAVELENGTH - ANGSTROMS 
FIG. 2. Spectral response of S-l and S-4 phototubes. 



The important characteristic of the new photosurface, its spectral 
sensitivity, is shown in Fig. 1. By contrast the same characteristic 
for the caesium-silver-oxygen S-l surface is also shown in Fig. 2. It 
should be borne in mind that these characteristics are typical of an 
average tube and that considerable variation from tube to tube may 
be expected. It is believed that the variations in the S-4 spectral 
response are proportionately much less than tnose encountered in the 
S-l surface. In addition to our own data, the data taken by a num- 
ber of investigators have recently been analyzed by us, and the curve 



382 



A. M. GLOVER AND A. R. MOORE 



Vol 46, No. 5 



shown is suggested as one which might be adopted as a typical 
standard. Variations in the position of the spectral maximum from 
4000 to 4500 A are commonly found with occasionally a peak at as short 
a wavelength as 3700 A being encountered, Based on data in greater 
quantity than previously available a curve giving the position of the 
maximum at 4200 A is believed typical. Little variation in the longer 



100 




WAVELENGTH -ANGSTROMS 

FIG. 3. Spectral response of S-4 surface with tungsten 
light at 2870 K. 



wavelength portion of the curve is encountered. Since the phototube 
will be commonly used in conjunction with an incandescent light 
source, the product of the spectral response curve of the tube and of a 
light source operating at 2870 K is shown in Fig. 3. This represents 
the effective spectral sensitivity of the tube as used in motion picture 
equipment. 

The sensitivity of a gas-filled phototube containing the S-4 photo- 
surface to a tungsten light source operating at a standard color tem- 
perature, 2870 K, may be made to vary considerably. For the pur- 



May, 1946 PHOTOTUBE FOR DYE-IMAGE SOUND TRACK 



383 



pose of introducing a replaceable phototube for the 868 into the 
motion picture industry a relatively high sensitivity is not required. 
The amount of argon gas added can be varied to make the resultant 
over- all sensitivity comparable to that of the 868 even though the 
energy output of the light source is not well adapted to the S-4 sur- 
face. The frequency response of the phototube is also a function of 
the gas pressure employed as is the breakdown voltage of the tube. 
With these factors in mind, an average gas amplification factor of 
three has been chosen. The gas amplification is less than that of the 



IVE RESPONSE -DECIBELS 
tit- + 
j> ^ KJ o rv) 
















































' " 1 U, 


^ 


s 


_ G/ 


^S AMP 


-IP 

. 




. 


















4 

- 


^ 

* r 


fc 






































p 


















































RELAT 

CD ( 















































































1000 ' 10000 

LIGHT-MODULATION FREQUENCY -CYCLES/SECOND 
FIG. 4. Frequency response of 1P37. 



100000 



868 and therefore the frequency response of the 1P37 is slightly better. 
The frequency response for 2 different gas amplification factors is 
shown in Fig. 4. These curves were obtained using a glow lamp modu- 
lated light source, 3 the data being corroborated by measurements with 
standard frequency sound track. 

It may be readily seen from the spectral characteristics that the 
blue-sensitive photosurface is more sensitive to variations in the tem- 
perature of the light source. This characteristic is believed to be its 
only point of inferiority when compared with the S-l surface. Data 
are shown in Fig. 5. Regulation of the light source voltage should be 
designed accordingly. It should be emphasized that the sensitivity 



384 



A. M. GLOVER AND A. R. MOORE 



Vol 46, No. 5 



figures quoted for any photosurface when expressed in microamperes 
per lumen vary with color temperature of the light source. Care 
should be taken to state on what basis data are quoted. 

Data giving the anode or output characteristic curves of the RCA 
1P37 phototube are shown in Fig. 6. These curves are comparable 
to those for the 868, thus permitting replacement of the latter tube 
by the 1P37 without modification of the circuit. 



lO-i 
















8 

6 

4 

2 
1- 

u 1.0- 














































VALUE 


S ADJUSTED TO GIVE 1 
OUTPUT AT 2870 K 


"QUAL 
















































/ 






LATIVE1 OUTPUT CURR 



r K * o. o, 






















s/ 














/ / 














/ / 












S 


/ 












/ , 


f 










86Q/ 
s 


' / 












/ 



TP37 










<> 

.06 
.04 

.02 
.01 




/- 














f 














I 














/ 














r 









































COLOR TEMP. OF TUNGSTEN LAMP SOURCE 

(DEGREES KELVIN) 

FIG. 5. Relative output of 868 and 1P37 as func- 
tion of light source color temperature. 

Prior to the war, one of the authors began work on the problem of 
making gas-filled tubes using the 5-4 surface. Considerable loss in 
sensitivity on use was encountered, and this conclusion was later 
stated in foreign articles on the subject. 4 Renewed efforts to elimi- 
nate or reduce this loss in sensitivity have met with considerable suc- 
cess. Life data for 2500 hr of continuous service are shown in Fig. 7. 
It is expected that life equal to that obtained with the 868 phototube 
will be obtained with the new tube. However, the life characteristic 



May, 1946 PHOTOTUBE FOR DYE-IMAGE SOUND TRACK 



385 




20 



40 60 80 

ANODE VOLTS 
FIG. 6. Anode characteristics of 1P37. 



100 



120 



1 SENSITIVITY -DECIBELS 

ill + + 

n ^ ro o ru 4^ 




SENSI 


TIVITY 


RELATI 


VE TO ' 


FHAT A - 


' HOI 


JRS 


























/X^ 


"^^ 




















\ 

x 


^ 


"S^ 


^cr / 
















^^^ 


- 


*^^^// 


V{T/AC 


CUR 


[NT =j 


LMICRC 


AMP. 


















-^. 5 





, 


> 

* 






















_l 8 

LJ 

o: 
-in 






















C 






















500 1000 1500 2000 25( 



LENGTH OF CONTINUOUS OPERATION . HOURS 
FIG. 7. Life data on 1P37. 



386 



A. M. GLOVER AND A. R. MOORE 



is somewhat different from that of the S-l in that the extremes of 
variation encountered each time the tube is operated are not so 
marked. As in other phototubes the life 
varies with the level of current drawn and 
the rated maximum average . current is 
set with this in mind. The currents 
shown are about those drawn with tubes 
of average sensitivity in a 35-mm optical 
system with the film in place. Greater 
currents may be drawn for brief inter- 
vals such as the time required for changing 
reels but limiting series resistance should 
always be included with a gas-filled photo- 
tube to prevent damage from extreme 
current in the case of voltage breakdown 
of the gas. 

Data taken with the IP 37 phototube, 
a photograph of which is shown as Fig. 
8, with dye image tracks, are the subject 
for other papers presented before the 
Society. 5 - 6 The structure of the 1P37 
has been dictated by the replacement prob- 
lem of the 868. The use of the photo- 
multiplier type RCA 931A in which 
the S-4 surface is included may also be 
of interest for developmental study. 

REFERENCES 




FIG. 8. 



Photograph of 
1P37. 



1 GORISCH, R., AND GORLICH, P.: "Reproduc- 
tion of Color Film Sound Records," /. Soc. 
Mot. Pict. Eng., 43, 3 (Sept., 1944), p. 206. 

- GLOVER, A. M., AND JANES, R. B., "A New High-Sensitivity Photosurface," 
Electronics, 13, (Aug., 1940), p. 26. 

3 KRUITHOF, A. A.: "Time Lag Phenomena in Gas-Filled Photoelectric Cells," 
Philips Tech. Rev., 4, (Feb., 1939), p. 48. 

4 SOMMER, A.: "The Influence of Illumination on the Fatigue of Photoelectric 
Cells," Elec. Eng., 17, (May, 1945), p. 504. 

5 DREW, R. O., AND JOHNSON, S. W. : "Preliminary Sound Recording Tests 
with Variable- Area Dye Tracks," J. Soc. Mot. Pict. Eng., 46, 5 (May, 1946), p. 
387. 

6 PHYFE, J. D.: "Behavior of a New Blue-Sensitive Phototube in Theater 
Sound Equipment," /. Soc. Mot. Pict. Eng., 46, 5 (May, 1946), p. 405 



PRELIMINARY SOUND RECORDING TESTS 
WITH VARIABLE-AREA DYE TRACKS* 



R. O. DREW AND S. W. JOHNSON** 

Summary. The introduction of new color emulsions to the industry makes it 
necessary for the sound engineer to investigate problems in connection with the record- 
ing and reproducing of quality tracks on this new medium. 

This paper is a discussion of the results of some preliminary tests made with 
variable-area tracks on Ansco reversible monopack color film. The tracks were proc- 
essed in exactly the same manner as for picture and therefore are of dye composition. 

Three specific methods of printing are discussed. Quantitative measurements of 
tracks containing modulated and constant-amplitude tones reproduced with the new 
gas-filled blue-sensitive and the ordinary red-sensitive phototubes are given. 

The problems of producing a satisfactory sound track on a con- 
ventional black-and-white film have been discussed in many papers 
in the JOURNAL of the Society during the past 18 years. Projection 
of pictures in color began with very limited use of systems in which 
alternate frames were projected through filters of complementary 
colors. Kodacolor, with its lenticulated film, opened the way to 
colored pictures from 16-mm film; while Technicolor, with its dye 
transfer process, gave us 3-color pictures in the theater. Sound 
could no doubt have been recorded on Kodacolor films, but the flut- 
ings of the base were a handicap and sound had not been widely ap- 
plied to 16-mm films when Kodacolor was superseded by the more 
practical and satisfactory Kodachrome. So far the problem of repro- 
ducing sound tracks recorded in other mediums than silver emulsion 
had not become serious, since the Technicolor process had provided 
for recording the sound in a silver emulsion, and then after clearing 
the remainder of the film, producing the colored picture. by trans- 
ferring dye colors to the clear gelatin. 

The Cinecolor process produces its blue color by iron toning, and the 
color thus formed works satisfactorily in sound systems designed for 
reproduction of silver tracks. 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Radio Corporation of America, RCA Victor Division, Indianapolis, Ind. 

387 



388 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



Reproduction of dye tracks assumed greater importance with the 
advent of multilayer films. Kodachrome, Ansco, and the new Koda- 
color film described by Mees 1 are examples. It is obviously desirable 
that the processing of the sound track be the same as for the picture. 
This means that the light absorption in the sound track must be pro- 
vided by the same dyes that produce the picture. The Report of the 



100 



80- 



60 



40 



20- 



SPECTRAL CHARACTERISTICS OF 
S4 AND SI SURFACES WITH TUNGSTEN LIGHT 




\ 



4000 



5000 



6000 



WAVE LENGTH - ANGSTROMS 
7000 6000 9000 10000 



1 1 000 



FIG. 1. Phototube characteristics of SI surface used in RCA Type 868 and 
S4 surface used in RCA Type 929 and IP37. 



Color Committee 2 in July, 1937, suggests that use of a phototube whose 
response, is confined to the visible spectrum would be desirable in giv- 
ing color technicians greater freedom in the choice of their coloring 
materials, but there is nothing in the report to indicate that a serious 
difficulty had already been encountered in efforts to reproduce from 
dye tracks with present optical systems. This difficulty seems first 
to have been called to the attention of our readers by Gorisch and 
Gorlich. 3 These authors point out that the dyes so far found suitable 
for the picture are quite transparent to infrared light while the 



May, 1946 



SOUND RECORDING TESTS 



389 



phototubes in general use are very sensitive to infrared light, their 
peak sensitivity with tungsten light being at about 8000 A, whereas 
7000 A represents the extreme of the visible. This relationship 
means that the phototube current is weakly modulated by the dye or 
dyes, but scratches and dirt modulate it almost as much as they 
would through clear film. The authors of that paper propose the sub- 
stitution of a blue-sensitive phototube which does not respond at all 



90- 



80 



TOP LAYtR 

MIDDLE LAYER.. 
BOTTOM LAYER- 



ALL LAYER! 



20 



X 



550 650 750 

WAVELENGTH I N ANGSTROMS 



650 



950 



FIG. 2. Dye characteristics of Ansco duplicating color film used in tests 

reported here. 



to infrared, and they describe a phototube with which they have 
worked with maximum response at about 5000 A. 

In 1944, Dr. A. M. Glover of RCA was approached by J. A. Ball of 
du Pont, who is Chairman of the SMPE Color Committee, to learn 
whether a phototube could be provided having the same spectral 
characteristics as the RCA 929, but with higher sensitivity, so that 
it could be substituted for the 868 caesium phototubes in existing re- 
producing systems, without the necessity of providing additional 
amplifier gain. The 929 phototube has its maximum sensitivity in 
the visible spectrum and has no response to infrared light. Inde- 



390 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



pendently and at almost the same time, the problem was presented 
to Dr. Glover by M. H. Sweet of Ansco. 



REPRODUCED 
REPRODUCED 

SILVER TRACK 



6000 O SILVER TRACK 




WITH BLUE SENSITIVE TUBE 
WITH RED SENSITIVE TUBE 







B 



FIG. 3. Relative output levels of color tracks. 

Fig. 1 shows the spectral characteristics of the 868 and 929 tubes. 
A new tube with similar cathode surface to the 929 tube and with 
identical spectral characteristic but with gas multiplication of the 
electron current has been developed, which has the desired sensitivity. 
It is designated as tube 1P37. The development of this tube is de- 



May, 1946 



SOUND RECORDING TESTS 



391 



BASE 



. . 

I. : .-;,.', ,,-.. -s--->.s'; /:' 



scribed in a paper by Glover and Moore. 4 The reader will find further 
discussion of this subject in the current Report of the Color Commit- 
tee. 6 The transparency of dyed gelatin to infrared light is general 
and not confined to certain dyes as is further indicated by a statement 
in the Eastman Wratten Filter catalogue which says that all of the 
gelatin filters are transparent to infrared. Fig. 2 shows the trans- 
mission characteristic of several dyes and of a complete track. 

To show that a phototube having the spectral characteristics of the 
1P37 tube would be more satisfactory than the infrared sensitive 868 
tube for reproducing sound recorded in dye tracks, did not call for ex- 
tensive tests. To learn how to get the best results from the dye 
track recordings and the 
blue-green sensitive 1P37 
tube is the next step, there- 
fore, and this paper is a 
report of a series of tests 
made with that object in 
view. 

In the figures we fre- 
quently refer to the 868 as 
the red-sensitive tube and 
the 1P37 as blue-sensitive, 
since it seemed to us that 
these descriptive terms would 
make the relationships easier 
for the reader to follow. It 
should be understood, however, that the 868 is one of a number 
of phototubes using what is designated as an S\ cathode surface, 
which is especially sensitive to infrared as well as to red light, and 
that the 1P37 and 929 phototubes use the S 4 surface which is 
highly sensitive to blue and to green light. 

All of our tests to date have been made on Ansco duplicating posi- 
tive color film. The general features of this film are described by 
Forrest and Wing 6 and by Duerr and Harsh. 7 The film, like other 
tripack systems, is arranged with a top layer emulsion sensitive only 
to blue, the next layer is a yellow filter followed by an emulsion sensi- 
tive to green, and finally, next to the base a layer sensitized to red. 
The emulsion layers, in the order named, upon being developed pro- 
duce yellow, magenta, and cyan or blue-green dyes. An arrangement 
for processing short pieces of film was set up in our laboratory and 




BASE 



DIFFUSION OF. LIGHT IN PRINT 



FIG. 4. Gradation of exposure through 
emulsion layers indicated by equal 
exposure contours. 



392 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



checks made which showed that our processing gave results substan- 
tially the same as those with films which we exposed and sent to Bing- 
hamton for processing. 

Although the final test for quality is actual listening to recordings 
of various classes of sound, there are 4 laboratory tests that have 
proved satisfactory for processing control in the case of variable-area 
silver tracks. These are volume level, high-frequency response, cross- 



SENSITOMETRIC CURVES 


FOR 2 LAYER TRACK 




EXPOSED WITH NO 186 FILTER IN SENSlTOMETER 
CVAN LArER FLASHED COMPLETELY WITH RED LIGHT 
BEFORE SENSlTOMETER EXPOSURE 


2. 5 














(.DENSITIES M 
WITH BLUE 
PHOTOTUBE 


EASURED 
SENSITIVE 




-~. 


LN 


, 








- f 
> o> i 

DENSITY 








\ 




CAM* 

\ 


|M, 


_i 
















^ 
























\ 


\ 








9 


04 9. 


34 9 


64 9 


94 0. 


LC 
24 


C.E 

54 


64 1 


14 u 


44 1 


* 

74 


.04 POSITIVE 



FIG. 5. Sensitometric curves for type lib sensito- 
metric exposure of 2 layers of Ansco reversible duplicat- 
ing color film. Densities were measured on micro- 
densitometer with tungsten light for the blue-sensitive 
phototube. 



modulation, and noise measurements. The usual volume level test 
is to measure the output of a 1000-cycle tone with full track modula- 
tion. Fig. 3 shows the outputs from dye tracks made in several ways. 
The 1000-cycle output from a silver track is taken as the reference 
or zero-db level. High-frequency response is judged from the level 
of a 6000-cycle tone as compared with the 1000-cycle tone. The 
cross-modulation test is that described by Baker and Robinson. 8 

Noise determinations consist in measuring the rms random noise 
output from an unmodulated biased track, and comparing the read- 



May, 1946 



SOUND RECORDING TESTS 



393 



SENSITOMETRIC CURVES 


FOR 3 LAY ER TRACK 


EXPOSED WITH NO.I86 FILTER IN SEN5ITOMETER 




\ 


^ 










-DEN 
BLU 


5ITIE 
E SE 


S M 

NSIT 


EASU 
IVE 


RED WITH 
PHOTO - 






\, 










TUBE 
















2. DENSITIES MEASURED WITH 
















RED SENSITIVE PHOTO- 
















TUBE 
















\ 






T 








1- 










\l. GAMMA 1.22 








*! 










\ 



























\ 



















GAM 

._ 


MA 


21 




\ 


V 


















L( 


2. 

. 

)G. E 


LL$ 


"*-* -~- 
L - 






J04 934 964 994 0.24 O.M 0.64 1.14 1.44 1 .74^2-.O4 POSITIVE 



FIG. 6. Sensitometric curves for type lib serisito- 
metric exposure of Ansco reversible duplicating color 
film. Densities were measured on microdensitometer 
with tungsten light for the blue-sensitive phototube 
and for the red-sensitive phototube. 



M ilffl 

r> ,~ UflMi 






0.0 

s_ 


R 70 iWMUfnWiiHTn 

pgB.::;! !!!!jj!||j!i[!!;l 
^ Ijjljjiijii!;;!;;:!!!! 

Ft nf I 


::: ::::$::j: : = PJU::S; \-\\S 




H 

55 

N 

Q 

0.4 


u :::::: 






Q_ 






^40 






400 450 500 550 600 650 
WAVELENGTH IN MILLIMICRONS 


700 



FIG. 7. Spectrophotometric transmission curves for Ansco yellow series 
filters Nos. 23, 24, and 25. 



394 



R. 0. DREW AND S. W. JOHNSON Vol 46, No. 5 



o 


BLUE SENSITIVE PHOTOTUBE 
2 LAYER TRACK- FILTERED LIGHT 
DIRECT POSITIVE MASTER 


\OOQf\j 

10 ' 


REFERENC 


E LEVEL 




6000 ^ 
















1- 
H 
-0 









-^^ 








-30 










\ 


1 


-40 










\ 


\l 


EK 


1372 D 

5 


RECT P 

2 


OSITIVE 




MASTI 

2 


:R DEN 

b 


5ITY 
3 



FIG. 8. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 2-layer tracks printed with tungsten light through 
one No. 25 Ansco filter after preliminary flashing with 
red light. The tracks were reproduced with the blue- 
sensitive phototube. 






2 


BLUE SENSITIVE PHOTOTUBE 
LAYER TRACK -FILTERED LIGHT 
PRINTED .MASTER 


1000 


r^> REFER 


INCE LEV 


EL 




6000 Oj 










ID 


?o 














1 

30* 










\ 




e 

o 



^ 

40 










\ 


\ / 


EK 
l 


J302 P 

5 


MINTED 

2 


MASTER 



DENS 
2 


TY 

5 


\ / 
3 



FIG. 9. Curves showing optimum printed master 
density and frequency response at 6000 cycles for 2- 
layer tracks printed with tungsten light through one 
No. 25 Ansco filter after preliminary flashing with red 
light. The EK 1302 printed master was printed from 
an original recorded EK 1372 negative of 2.65 density. 
The color tracks were reproduced with a blue-sensitive 
phototube. 



May, 1946 



SOUND RECORDING TESTS 



395 



ing with the 1000-cycle full modulation output. The unmodulated 
biased track consists of 2 clear strips 0.003-in. wide with the remainder 
of the scanned area opaque. Film noise is caused by any photographic 
irregularities or graininess which may be present, and by scratches 
and specks of dirt. Such dirt specks as appear over the narrow clear 
areas produce fairly loud noises. Dirt in other areas causes noise in 
proportion to the light transmitted, so that low densities mean noisy 
films. The ratio of signal to noise, or relative noise level, is not greatly 





BLUE SENSITIVE 
3 LAYER TRACK - 
DIRECT POSIT IV 


PHOTOTUBE 
WHITE LIGHT 
E MASTER 


lOOOOj 
10 


REFERENC 


E LEVEL 




eooo'Vi 


















8 




_ 


. 


~^^ 






J 








X 


\ 


/ 


t , 










\ 


/ 


EK 

1. 


1372 D 


IRECT 2 


'OSITIVE 
.0 


MASTE 


!* DEN 


/ 

S.TY 3 



FIG. 10. Curves showing optimum direct-positive 
master density and response at 6000 cycles for 3-layer 
track printed with unfiltered tungsten light. The 
tracks were reproduced with the blue-sensitive photo- 
tube. 



affected by some light absorption in the clear areas, since this re- 
duces intensities of both signal and noise. Low densities in the dark 
areas, on the other hand, raise the absolute magnitude of the noise and 
at the same time reduce that of the signal, and are therefore doubly 
injurious. Graininess is not likely to be a serious source of noise in 
dye tracks developed by reversal. 

Both the high-frequency loss and the cross-modulation are the re- 
sult of diffusion of light within the emulsion, which causes the ex- 
posure to spread outside the boundaries of the area where exposure 
is desired. This takes place especially in the lower layers of the emul- 
sion. Fig. 4 illustrates the effect of light diffusion in the emulsion 



396 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



BLUE SENSITIVE PHOTOTUBE 
3 LAYER TRACK- WHITE LIGHT 
, DIRECT POSITIVE MASTER 


10000 
-10 


jREFERENC 


E LEVEL 


eoooaj 




















o 

Q 
1 

-20 







"^^ 


"N 


/ 




- 
o 
a. 
t- 

3 









\ 


J 






* 
-40 














EK 13 

i. 


57 DIP 

5 


ECT PC 

2 


)STIVE 




MASTER DEN: 

2.5 


>ITY 

3 



FIG. 11. Curves showing direct-positive master 
density and frequency response at 6000 cycles for 3-layer 
tracks printed with unfiltered tungsten light from EK 
1357 master. The color tracks were reproduced with 
the blue-sensitive phototube. 



BLUE SEN STIVE PHOTOTUBE 
3 LAYER TRACKS-FILTERED LIGHT 
DIRECT POSITIVE MASTER 


1000 ry. 
10 


REFEREIV 


CE LEVE 


. 

eooooj 




















o 
o 

"'O ' 














H 


~^ 


^ 




x 


^ 


- 



O 

* 






Y 


V 






EK 1 


372 Dl 

5 


RECT P 

2 


OSITIVE 




MASTf 

2 


:R DEN 

5 


SITY 



FIG. 12. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3-layer tracks printed with tungsten light through 
3 thicknesses of No. 25 Ansco niters from EK 1372 
masters. The color tracks were reproduced with the 
blue-sensitive phototube. 



May, 1946 



SOUND RECORDING TESTS 



397 



layer of a film being printed from a relatively sharp negative of high 
density. In recording and printing silver sound tracks, a major im- 
provement is made by confining the image close to the surface by ex- 
posing with ultraviolet light. In high-contrast negatives and non- 
reversal prints exposed to produce high densities, the spread of expo- 
sure results in outward shifting of the boundary of black areas, or 
what we call "image spread." It is evident that image spread in a 
negative offsets image spread in a print. This is because the image 





BLUE SENSITIVE PHOTOTUBE 
3 LAYER TRACK FILTERED LIGHT 
DIRECT, POSITIVE MASTER 


1000 a 


REFEREN< 


:E LEVEL 


6000^ 
















V 


^ 






^-~ 






- V 

D 

a. 
I 




\ 


1 








^r 




\l 










EF 
i 


( 1375 

5 


DIRECT 

2 


POSITI 




VE MA 

2 


5TER D 

5 


iNSITY 

3. 



FIG. 13. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3-layer tracks printed with tungsten light through 
3 thicknesses of Ansco No.' 25 filter from EK 1357 
master. The color tracks were reproduced with the 
blue-sensitive phototube. 



spread in the negative tends to make undersized the black areas in a 
nonreversal print. The cross-modulation method of processing control 
consists in choosing a combination of negative and print densities at 
which the image spread in the negative just balances out that in the 
print. For this balance the cross-modulation recording is a delicate 
test, and conditions that give minimum cross-modulation mean mini- 
mum distortion and the best sounding prints. Commercially, prints 
are considered satisfactory if cross-modulation is 30 db below signal. 
Such balancing of image spread does not eliminate high-frequency 



398 



R. O. DREW AND W. S. JOHNSON 



Vol 46, No. 5 



losses, which are substantially the same over a range of densities 
on both sides of that which gives best balance. 

Since the Ansco Color film with which we worked is processed by 
reversal, the effect of diffusion of light in the emulsion is to eat into the 
area which should be opaque, and cause image contraction instead of 
image spread. To neutralize this image contraction, the black area 
of the master (we cannot in this case call it a negative) has to be over- 
size, thus protecting a slightly larger area than that which is to be 



RED SENSITIVE PHOTOTUBE 


3 LAYER TRACK -WHITE LIGHT 


DIRECT P 


OSITIVE MAST 


ER 




10000. 


REFEREN 


CE LEVEL 
















6000O) 








-10 




^==r: 











o 






CROSS MO 


DULATION 


^^^^ 




1 








CURVE 


^^ 


v. 


8 














40 














EK 1 


372 DIP 


ECT PC 


35ITIVE 


MASTE 


R DEr 


vlSlTY 


1.5 2.0 2.5 3. 



FIG. 14. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3 -layer tracks printed with unfiltered tungsten 
light. The tracks were reproduced with the red-sensitive 
phototube. 



cleared of dye. With reversal prints, the plotted cross-modulation 
measurements show the same kind of V curves as are shown in the 
case of negative-to -positive prints, but the distortions on opposite 
sides of the minimum point are of opposite sign. Thus, if the ab- 
scissae are master (or negative) density, the branch of the curve to 
the right of minimum means undersize opaque image in the case of 
ordinary prints, but indicates oversized image in the case of reversal 
prints. 

There are also more exacting requirements in the case of the rever- 
sal print. In all variable-area sound prints, it is important to keep 
the transparent areas clear, and this dalls for complete exposure of 



May, 1946 



SOUND RECORDING TESTS' 



399 



the emulsions. On the other hand, since all of the density obtainable 
is wanted in the dark areas, this part of the track must be almost 
completely protected from exposure. 

The ideal printing master is a direct recorded positive. Most of our 
tests were with direct positives, but we have also tried printed mas- 
ters. In either case, high densities and some image spread are wanted. 

Since in silver prints best high-frequency response and density 
tolerances for good cross-modulation cancellation had been obtained 



RED SENSITIVE PHOTOTUBE 
3 LAYER TRACK - WHITE LIGHT 


, DIRECT F 


>OSITIV 


: MAST 


ER 




3 .oooa 
-10 


j REFERENC 


E LEVEL 

. 

^^ 


' 


: ^ 


^^ 



\^^ 


8 










CROSS MO 
CURVE 


JULATION 


Q. 

- 

O 














O 
O 














EK 

i. 


357 DIP 
5 


IECT P( 

a 


)5ITIVE 




MA5TI 

2 


:R DEN 

5 


SITY 

3. 



FIG. 15. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3-layer tracks printed with unfiltered tungsten 
light. The tracks were reproduced with the red-sensitive 
phototube. 



by confining the exposure as near the surface as possible, it was an- 
ticipated that similar benefits would result from using only the surface 
layers of the tripack. This can be readily accomplished by clearing 
the bottom layer by an auxiliary exposure to red light, leaving only 
the dyes in the 2 top layers to produce the modulation. Tracks so 
printed are here designated as "2-layer tracks," in distinction to "3- 
layer tracks" wherein no flashing exposure is employed, and all 3 
dyes are produced in the dark areas. 

A slight advantage of the kind expected was found for the 2-layer 
track, but it is not surprising that the margin of advantage is small. 
The diffusion of printing light under edges of the black area extends 



400 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



farthest in the bottom layer thus shading off the edges of the image, 
but since this takes place principally in the cyan-dyed layer, and the 
cyan dye is transparent to the blue and green light to which the 1P37 
tube is most sensitive, the presence or absence of some of the cyan 
dye makes comparatively little difference in the phototube current. 
Such difference as it does make results from the fact that the cyan dye 
is not 100 per cent transparent in this range. Figs. 5 and 6 show the 
relation of density to exposure in 2- and 3-layer tracks, the densities 
being calculated from phototube currents. 



RED SENSITIVE PHOTOTUBE 
3 LAYER TRACK- FILTERED LIGHT 
DIRECT POSTIVE MASTER 
J ' 


lOOO'V 


i REFERE 


JCE LEVEL 


60000. 








S 

1 










^\ 




CR< 


SS MODU 


-ATION C 


JRVE 


I 

D 
O 












S 


<? 


O 

* 














EK 

i. 


1372 Dl 

5 


RECT P 

2. 


05ITIVE 




MASTE 

2 


R DEN 

5 


>ITY 

3. 



FIG. 16. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3-layer tracks printed with tungsten light through 3 
thicknesses of Ansco No. 25 filters. The tracks were 
reproduced with the red-sensitive phototube. 

Some trials were made of single-layer tracks, using only the yellow 
dye in the top layer to do the modulating. If the 1P37 tube were 
sensitive only in the blue, the single- layer track would be logical, but 
it is sensitive far into the green, and hence to absorb all of the light 
to which the tube is sensitive the yellow dye must be supplemented 
by the magenta (minus green) dye. 

Since the color film prints made with white light showed a blue tint 
in the dense areas, trial was made of yellow filters in the printing light. 
Fig. 7 shows the transmission of the Ansco No. 25 filter of which 3 
thicknesses were used in printing the 3-layer tracks, and one thick- 



May, 1946 



SOUND RECORDING TESTS 



401 



ness in printing the 2-layer tracks. The use of the yellow filter per- 
mits relatively more of the yellow dye of the top layer to be formed, 
and there is less image contraction because undercutting is least in 
the top layer. As shown in Figs. 10 to 13, best cancellation occurs at 
a lower master density than in the case of color prints made with 
white light. 

Figs. 8 and 9 indicate satisfactory cross-modulation cancellation 
for 2-layer tracks printed from either a direct positive or a printed 



RED SENSITIVE PHOTOTUBE 
3 LAYER TRACK - FILTERED LIGHT 
DIRECT POSITIVE MASTER 




-40 



EK 1357 DIRECT POSITIVE MASTER DENSITY 

1.5 2.0 2.5 



FIG. 17. Curves showing optimum direct-positive 
master density and frequency response at 6000 cycles 
for 3 -layer track printed with tungsten light through 
3 thicknesses of No. 25 Ansco filter. The tracks were 
reproduced with the red-sensitive phototube. 



master. The 6000-cycle output is only 3.5 db below the 1000-cycle 
output. 

Figs. 10 to 13 show that filtered printing light gives cancellation 
at lower master densities than does white light, and slightly increased 
density tolerance for 30-db cancellation. Also, direct-positive mas- 
ters recorded on the fine-grain, low-spread E.K. 1372 give slightly 
more 6000-cycle output, and greater density tolerance for 30-db 
cancellation than is given by masters recorded on the somewhat 
coarser grained E.K. 1357. The 3-layer tracks of Figs. 10 to 13 
give about 1.5 db less 6000-cycle response than the 2-layer tracks. 



402 



R. O. DREW AND S. W. JOHNSON 



Vol 46, No. 5 



Both give about the same 1000-cycle response, which is 4 db below 
that of a silver track. 

Apart from determining optimum recording conditions for a dye 
track used with a blue-green sensitive tube, and demonstrating that 
the combination works well, the purpose of our investigation was, 
using 1P37 tube, to compare the dye track with a silver track, and 
also, using the dye track, to compare the results using the blue- 
sensitive 1P37 with those obtained with the red-sensitive 868 tube. 







3 LAYER TRACK 






20 I 


















2 
? 

in 






X 


X^ 


REPRODUCED 

RED SENSITIV 
PHOTOTUBE 


WITH 

E 


-3Oo 

1 




/ 


/ 










g 

3 


/ 


/ 












*EJ 

a 

$ 


/ 




^ 


-"" 


REPRODUCED WITH 
BLUE SENSITIVE 
PHOTOTUBE 


_i 

ui 
to 




/ 


^ 














NU 
) i 


YlBER OF 
)0 2 


RUNS 
00 . 3 


THRO 
9O 4 


PROJEC 
DO 5 


TOR 
00 6 


00 



FIG. 18. Curves showing relation between signal 
to noise for 3 -layer biased unmodulated track and 
number of runs through a theater-type projector. 



Fig. 2 shows the comparison of levels. It is obviously not fair to the 
868 tube to make the comparison with any but the 3-layer track. In 
the 2-layer track as we made it, the cyan-dye layer was cleared, thus 
removing the only dye that effectively modulates red light. Using the 
3-layer track, the 868 tube is still at a disadvantage, not only in the 
transparency of the dyes to the infrared light which produces most of 
the phototube current of the 868 tube, but also in the arrangement of 
the layers which puts the cyan-dye layer at the bottom, or farthest 
from the surface where the printing light is applied, and therefore 
where it gets the poorest printing conditions. 



May, 1946 SOUND RECORDING TESTS 403 

The low maximum density reached when the 868 phototube is used 
is indicated in Fig. 6. Figs. 3 and 14 to 17 show the results of the 
tests of output and cross-modulation with the 868 tube. It appears 
that no practically obtainable master density gives enough image 
spread to offset the image shrinkage in the color film, although an 
approach toward balance appears in Fig. 17. 

In both the 2-layer and 3-layer tracks reproduced with the 1P37 
phototube, the 1000-cycle outputs were about 4 db below that of a 
silver track. Ground noise with fresh film was likewise about 4 db 
below that of the silver track, each of these tracks showing a noise 
level 41 db below its own 1000-cycle output. The 2-layer track 
was then run 200 times through a projector, and it was found that 
both the silver and the dye tracks increased in absolute noise output 
power by substantially equal amounts. In another test the increase 
of noise from a 3-layer dye track owing to wear was determined. 
Fig. 18 shows the noise increase as measured with an 868 and a 1P37 
phototube. The rapid increase in noise in the case of the 868 tube 
is what would be expected in view of the low density of the dark por- 
tions of the track. 

Summarizing the results of our tests, good high-frequency response 
and low distortion are obtainable with dye tracks and 1P37 photo- 
tubes, with levels slightly below those from silver tracks, and with 
signal-to-noise ratios practically the same as with silver tracks. 

Two-layer tracks have slightly superior performance as compared 
with 3-layer tracks. Good tracks can be made from either direct 
positives or printed masters, provided a good negative is available 
for making the master. Masters should have higher densities and 
more image spread for printing color tracks than for making standard 
silver tracks, and the control for minimum distortion follows the usual 
cross-modulation test procedure. 

REFERENCES 

1 MEES, C. E. K. : "Direct Processes for Making Photographic Prints in Color," 
/. Soc. Mot. Pict. Eng., 42, 4 (Apr., 1944), p. 230. 

2 REPORT OF THE COLOR COMMITTEE: J. Soc. Mot. Pict. Eng. t XXIX, 1 (July, 
1937), p. 54. 

3 GORISCH, R., AND GORLICH, P.: "Reproduction of Color Film Sound Rec- 
ords," /. Soc. Mot. Pict. Eng., 43, 3 (Sept., 1944), p. 206. 

4 GLOVER, A. M., AND MOORE, A. R.: "A Phototube for Dye Image Sound 
Track," /. Soc. Mot. Pict. Eng., 46, 5 (May, 1946), p. 379. 

6 REPORT OF THE COLOR COMMITTEE: /. Soc. Mot. Pit. Eng., 45, 6 (Dec., 1945), 
p. 397. 



404 R. O. DREW AND S. W. JOHNSON 

6 FORREST, J. L., AND WING, F. M.: "The New Agfacolor Process," /. Soc. 
Mot. Pict. Eng., XXIX, 3 (Sept., 1937), p. 248. 

7 DUERR, H. H., AND HARSH, H. C. : "Ansco Color for Professional Motion 
Pictures," /. Soc. Mot. Pict. Eng., 46, 5 (May, 1946), p. 357. 

8 BAKER, J. O., AND ROBINSON, D. H.: "Modulated High-Frequency Record- 
ing as a Means of Determining Conditions for Optimal Processing," /. Soc. Mot. 
Pict. Eng., XXX, 1 (Jan., 1938), p. 3. 



BEHAVIOR OF A NEW BLUE- SENSITIVE PHOTOTUBE IN 
THEATER SOUND EQUIPMENT* 

J. D. PHYFE** 

Summary. A new phototube designed to provide optimum performance when 
used in reproducers with films having standard black-and-white silver sound tracks, 
or with color films having either dye or edge-treated sound tracks, has been developed 
in the RCA laboratories. 

Some results of laboratory tests and field observations when the new phototube is 
substituted for the standard red-sensitive 86 8 -type phototube are discussed. 

Need has recently arisen for a phototube that could be interchanged 
with the type commonly used in theater sound heads. The necessity 
is the result of an apparent trend toward an increasing use of color in 
motion picture film productions. Also, there is the possibility that a 
larger percentage of these color films will have dye sound tracks in- 
stead of the usual silver tracks. The desired tube should be inter- 
changeable with the red-sensitive type, therefore, and must perform 
as well when used with films that have the regular silver sound tracks. 

Such a phototube has been developed in anticipation of these needs, 
and is known commercially as the type IP 37. To date the observed 
performance of this tube indicates that it accommodates this change- 
over very well. Mechanically and electrically the 1P37 is inter- 
changeable with the type 868 phototube which has been used in RCA 
theater sound equipment for more than a decade. 

In order to evaluate the merit of this new blue-sensitive phototube 
for theater use, comparisons of performance have been made in the 
laboratory with the type 868 red-sensitive tube. In addition a num- 
ber of the blue-sensitive phototubes have recently been distributed 
to various theaters throughout the country with instructions to use 
them in place of the 868 tube, and report their relative behavior. 

Too short a time has elapsed since these phototubes were placed in 
the field for observation, so reports on performance are not complete 
at present. 

Laboratory tests have been made, however, to determine how well 
they would function in theater sound heads with standard release 



* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Radio Corporation of America, RCA Victor Division, Indianapolis, Ind. 



405 



406 J. D. PHYFE Vol 46, No. 5 

prints, and if there were any adverse operating characteristics which 
might preclude their being substituted for the 868 tube. Some of the 
characteristics investigated were: 

(1) Relative gain as compared to the 868 tube when used with films having the 
usual silver sound tracks. 

(2) Relative hiss level and microphonics. 
(5) lonization or glow point. 

(4} Relative distortion. 

(5) Relative hum level when used with raw (unfiltered) a-c on the exciter lamp. 

(6) Gain variation with changes in exciter lamp brilliancy owing to line voltage 
fluctuations. 

(7) Ease of balancing output between sound heads by varying the anode volt- 
age. 

(8) frequency response relative to the 868 tube. 

(9} Refocusing of the optical system because of different spectral sensitivities 
of blue-sensitive and red-sensitive phototubes when interchanged. 

Results of Laboratory Tests. The results of these tests are re- 
ported in the order in which they were enumerated above : 

(1) Gain Check. This was made by running a 1000-cycle loop of 
standard silver sound track through a regular theater sound head and 
amplifier channel, and noting the levels.. Both types of phototubes 
were used. A sufficient number of both types were checked to repre- 
sent an average cross section of sensitivity variation. Exciter lamp 
and phototube anode voltages were held constant at 8 l / 2 and 75 v, 
respectively. 

Results of this check showed that the sensitivity of the 2 types of 
phototubes is practically the same with a slight superiority of the 
1P37 tube. It was noted that the 1P37 tube showed somewhat less 
variation in output from one tube to another than tube 868. 

(2) Hiss Level and Microphonics. This test was made by checking 
the relative outputs of a group of red-sensitive and blue-sensitive 
phototubes using a 1000-cycle film loop as a signal source. The film 
was then removed, the amplifier gain was raised to approximately 130 
db and the hiss level was measured. The hiss level was checked both 
on an output meter and a sound-level meter used in conjunction with 
a standard 2- way theater loudspeaker system. 

Microphonics was then checked by starting the projector motor and 
noting the output level, as was done in measuring hiss. 

The results of these tests revealed no apparent difference in hiss or 
microphonics when the blue-sensitive phototube was used. 

(3) lonization or Glow Point. No changes in voltage supply are 



May, 1946 BEHAVIOR OF BLUE-SENSITIVE PHOTOTUBE 



407 



necessary when changing from the type 868 phototube to the IP 37 
type. The same maximum supply voltage limitations exist for 
both types. 

(4) Distortion. A 
stant-frequency film 



con- 
having 

80 per cent modulation was 
run through a standard theater 
sound head and theater ampli- 
fier channel using both types . 
of phototubes, and the rms 
harmonic distortion measured 
by means of a distortion fac- 
tor meter. The measured dis- 
tortion was found identical for 
the 1P37 tube and the 868. 

(5) Relative Hum Level. 
Using a-c on the filament of a 
standard 10- v, 7.5-amp exciter 
lamp, the 1P37 tube showed 4 
db more hum when the exciter 



EXCITER 
LAMP 
VOLTS 
A.C. 


OUTPUT WITH 
1000 CYCLE FILM 


HUM OUTPUT 


HUM OUTPUT 
BELOW 1000 CYCLE 
FILM OUTPUT 


IP37 


868 


IP37 


868 


IP37 


868 


75 


17.7 


19.1 


-04 


-3.4 


18.1 


22.5 


8.0 


20.0 


20.4 


+2.0 


-1.6 


18.0 


220 


8.5 


21.8 


21.8 


+3.8 


-0.2 


18.0 


22.0 


9.0 


23.3 


22.7 


+5.7 


+ 1.1 


17.6 


21.6 


9.5 


24.8 


23.8 


+74 


+2.6 


174 


21.2 


10. 


26.5 


24.9 


+10.0 


+4.0 


1*6.5 


20.9 



FIG. 1. Relative film and hum output 
variations in db IP 37 versus 868. Film 
removed for hum output measurements. 



T 

4-.A 














IP37 










^ 


&Z 


**? 


868 


(/) 

id -4 




^? 


^ 


OUTPUT 


WITH 1 


000 CYC 


LE FILM 
















h 

D -16 














P37 










/ 


/ 




-20 
-24 
-28 






^ 


^^RL 


HUM 

M REMC 

_^, 


OUTPUT 
)VED^- 


868 




^ 


^ 


^ 










>^ 












75 8 85 9 9.-S iu 
EXCITER LAMP VOLTS A.C. 



FIG. 2. Relative signal-to-hum response. 



phototube and 8.8 
film. This shows a 



lamp voltage was adjusted to a 
normal operating value of 8.5 
v. This was based on equal 
signal outputs for both types 
of phototubes using a 1000- 
cycle film loop. Figs. 1 and 
2 show relative signal output 
and hum levels between the 
1P37 tube and the 868 when 
the exciter lamp voltage was 
varied between 7.5 and 10 v. 
Hum level was measured by 
removing the film after output 
measurements were taken. 

(6) Gain Variation with 
Changes in Exciter Lamp Volt- 
age. Referring to Figs. 1 
and 2, a gain change of 5.8 
db is observed for the 868 
db for the 1P37 tube, using a 1000-cycle 
3-db increase in gain variation for the 1P37 



408 J. D. PHYFE 

tube when the exciter lamp voltage is varied between 7.5 and 
10 v. 

(7) Balancing Sound Head Outputs by Adjusting Anode Potential 
of Phototube. The adjustments used for balance of the sound head 
outputs are the same for both the 1P37 tube and the 868. These are 
made by control of anode supply voltage. 

(8) Optical System Focus When Using Both Types of Phototube 
A standard lY^mil slit image optical system was focused for maxi- 
mum output using a 7000-cycle loop of film and an 868 red-sensitive 
phototube. Output readings were then taken for both types of photo- 
tubes. No observable increase in output was obtainable with the 
IP 37 tube by refocusing the optical system. This test was then re- 
peated using a 9000-cycle loop of film. The results were identical. 

This observation seems to substantiate the assumption that there 
is no need to ref ocus a standard 1 V 4 -mil slit image optical system when 
the 1P37 tube is used in place of the 868. 

(9) Relative Frequency Response. For this check a standard thea- 
ter reproducing channel having an optical system with a IVVmil 
slit image was adjusted for maximum focus using a 9000-cycle 
loop. No low-pass filter was used since this might have had the ef- 
fect of making relative output at the higher frequencies. Frequency 
runs were then made using a calibrated test film which included 31 
different frequencies between 30 and 9000 cycles. Response measure- 
ments revealed no difference in relative frequency response when the 
1P37 tube was substituted for the 868. 

Conclusion. Reports from the field are awaited in order to better 
judge the seriousness of the increased hum, and greater variation 
in signal level with changes in exciter lamp voltage obtained with 
the 1P37 tube. No attempt will be made at this time to evaluate the 
seriousness of these 2 points. 

It is felt, however, that the increased hum might be satisfactorily 
compensated by modification of the 120-cycle hum filter in those in- 
stallations which operate with raw a-c on the exciter lamp. 

For those installations which operate with d-c exciter lamps and 
which have some form of regulation of the exciter lamp voltage, it 
does not appear that the latter point would become a problem. 

In the meantime, however, the tubes that have been substituted 
are working very well, indicating that no differences in operating 
characteristics have been observed, or else are not of sufficient mag- 
nitude to justify an immediate report. 



ELECTRONIC SHUTTER TESTERS 5 
R. F. REDEMSKE** 



Summary. The following paper, reprinted from Electronics, describes a photo- 
electric system which feeds a bank of stylii producing on Teledeltos paper a recording 
of camera shutter-opening area plotted against time, for testing both iris and focal- 
plane shutters. Direct-indicating accessory shows per cent deviation from rated 
shutter speed. 

The editors of the Journal feel that the methods and equipment outlined here will 
be of timely interest to members in the motion picture industry and allied fields. 

Military demands for large numbers of precision aerial cameras 
during World War II dictated an accelerated mass production pro- 
gram. With this program arose the need for a shutter tester of high 
accuracy, applicable to production-line use. Previous shutter- 
testing methods, employing both photographic and electronic tech- 
niques, were well suited to laboratory use but hardly applicable to 
production requirements. 

One requirement of the new tester was that it be capable of yielding 
a permanent record of shutter characteristics for both iris-type 
(between-the-lens) shutters and focal-plane shutters. In the case 
of the iris type, the record should indicate both speed and efficiency, 
while for the focal-plane type the record should show the shutter 
speed at 3 points : near the beginning, center, and end of the curtain 
travel. This instrument must also have an auxiliary time-measuring 
circuit for testing the K-19 night photo camera. Another require- 
ment was a visual indicating shutter tester for both iris and focal- 
plane shutters. This article describes shutter testers that evolved 
from this development program and outlines the factors influencing 
the design. 

In considering the requirement that data from the recording 
shutter tester be in the form of a permanent record, it was deemed im- 
practical to use any method involving a photographic process. This 

* Reprinted from Electronics, 19, 2 (Feb., 1946), p. 128. 

** Project Engineer, Fairchild Camera and Instrument Corporation, Jamaica, 
N. Y. 

409 



410 R. F. REDEMSKE Vol 46, No. 5 

conclusion was reached because the time consumed in the develop- 
ment, fixing, washing, and drying sequence would be prohibitive, and 
because a setup involving a darkroom is inconvenient. This elimi- 
nates photographic methods of shutter testing, as well as electronic 
methods that employ a recording string-oscillograph. 

Other possible recording means were studied, and as a result of this 
investigation it was decided to use Teledeltos recording paper, which 
has the property of carbonizing and forming a dark line when a stylus 
energized with sufficient voltage is drawn against it. 

Recording Shutter Tester. The best recorded data to provide 
speed and efficiency figures for an iris shutter is the curve of shutter 
opening area versus time. A typical curve is shown in Fig. 1 A. The 
advantage of this type of presentation is that it provides sufficient 
data to rate shutter speed on any of the several bases. To reproduce 



IRIS SHUTTER 
CHARACTERISTIC 


IRIS SHUTTER TESTER 
PRESENTATION 


FOCAL PLANE SHUTTER 
TESTER PRESENTATION 

NEAR START SPEED 
(STYLUS 1) NEAR END 
H (STYLUS 3) 

AT^NTEF I" -\ 


(A, 


,B, T1ME 


(STYLUS?! 
(C) ~~ 



FIG. 1. Examples of records obtained with electronic shutter tester for iris 
and focal-plane shutters. 

this curve on Teledeltos paper, it was decided to move the paper 
under 10 stylii equally spaced in a straight line and controlled in- 
dividually, to give the result shown in Fig. IB. 

With this setup the focal-plane shutter tester could use three of 
the stylii to show the speed at 3 curtain positions, giving a record like 
that in Fig. 1C. The recorded lines would not necessarily appear in 
sequence, as it is possible for the recording drum to make more than 
one revolution between traces. 

The paper was moved beneath the stylii by mounting it on a cylin- 
drical drum of 15-in. circumference, driven by a synchronous motor. 
The stylii cannot be allowed to rest too long on the paper with the 
drum rotating as they wear marks which can be confused with the 
signal traces. To avoid this trouble, the stylii arms are held off the 
paper by a spring system. Just before the shutter is tripped, a 
switch is thrown to energize* a solenoid which overcomes the spring 
and contacts the stylii with the paper. Two drum speeds are pro- 
vided. This is necessary to prevent the traces for low shutter speeds 



May, 1946 ELECTRONIC SHUTTER TESTERS 411 

from taking up more than one revolution and overlapping, and still 
be able to provide a long enough trace for accurate measurement at 
high shutter speeds. Knowing the drum speed and circumference, 
the lengths of trace for various shutter speeds can be calculated. 
Typical values are given in Table 1 . 

TABLE 1 

Relation of Shutter Speeds to Trace Lengths 

Drum Speed 
Shutter 

Speed 1200 rpm 2400 rpm 

(Sec) (In.) (In.) 

V 12 

V0 6 

Vioo 3 

Vo .... 3 

VMO 1.2 



The shutter speed tolerance of 10 per cent can readily be detected 
on this basis. The system necessary to accomplish this presentation 
of data on Teledeltos is outlined in block diagram form in Fig. 2. 

Iris Shutter Optic and Phototube Unit. In the case of the iris 
shutter, the first step necessary to achieve the end result shown in 
Fig. IB is to convert the shutter opening area into a corresponding 
electrical voltage. This was accomplished by the photoelectric 
arrangement of Fig. 3. A light source consisting of a tungsten- 
filament bulb in a reflector is placed a considerable distance above 
the shutter. This applies uniform illumination with approximately 
parallel light rays to one side of the shutter. 

It was found necessary to operate the lamp from a d-c source to 
avoid ripple pickup by the phototube. A lens system on the other 
side is adjusted to focus an image of the light source on a piece of 
diffusing material which in turn reflects light to the phototube. The 
shutter then acts as a variable iris controlling the amount of light 
received by the phototube. This proved to be the best way to main- 
tain a linear relationship between shutter opening area and phototube 
output current. The voltage output from the phototube load 
resistor when the shutter is operated is therefore varying as in Fig. 
1A. The system that translates this voltage pulse into lines on the 
Teledeltos paper must have a frequency response range such that no 
distortion is introduced. 



412 



R. F. REDEMSKE 



Vol 46, No. 5 



To consider the frequency response requirements, refer to Fig 4. 
To transmit without distortion an isolated trapezoidal pulse in time 
a general rule is that the circuit must be flat in response up to a fre- 
quency/determined by 1/7", where T is the time duration of the slope. 
The pulse obtained from the phototube approaches the shape of Fig. 
4. Assuming that the highest shutter speed to be encountered 
would be Yiooo sec and assuming that the slope duration might be 10' 



TYPE OF 
SHUTTER 



NIGHT 
PHOTO 




AMP 









STYLUS 
AMPLIFIERS 

, STYLI 

NIGHT ' 
PHOTO 













IRIS 




AMP 




rt~ \ 





IRIS-FOCAL PLANE 
TEST SELECTOR 



FIG. 2. Block diagram of recording shutter tester. 

per cent of this, the upper frequency limit/ is 10,000 cps. Circuit 
constants to meet this frequency requirement are 

/ = V 2 x RC (1) 

where / is the frequency where response is 3 db down, R is load re- 
sistance, and C is shunt capacitance. 

In designing the photoelectric circuit,- it is desirable to use the 
maximum permissible value of load resistance to achieve the maxi- 
mum signal voltage, and thus reduce the amplifier gain requirements. 
However, as the phototube is built into the shutter test jig, it must be 



May, 1946 



ELECTRONIC SHUTTER TESTERS 



413 



X 



^un 



\ LIGHT 
SOURCE 



SHUTTER 



separated from the main amplifier-recorder rack by some 5 ft of 
cable. This introduces a high value of shunt capacitance that 
severely limits the value of phototube load resistance when Eq (Jf) 
is considered. It was accordingly decided to use a cathode follower 
as an impedance changer at the 
phototube to reduce the shunt 
capacitance across the load re- 
sistor to a minimum. The cir- 
cuit is shown in Fig. 5. By 
locating the 6SN7 very close to 
the 929, it was possible to use 1 .3 
megohms for R L . 

Focal-Plane Optic and Photo- 
tube Unit. To measure the 
speed of a focal-plane shutter, it 
is necessary to measure the 
time required for the shutter 
slit to pass a point. The ar- 
rangement in Fig. 6 was de- 
signed to do this. This setup 
provides a source of illumination to the top of the shutter while 
the shutter opening is passed over a narrow slot parallel with it 
and very closely under it. The light source is restricted with a slot 
to make the light applied to the shutter more nearly parallel. During 
the coincidence of the openings, light will fall on the cathode of the 
phototube. The output of the phototube will then also be in the 

form of a trapezoidal pulse. The 



LENS 



DIFFUSING 
PLATE 




PHOTOTUBE 



FIG. 3. Optical system of shutter 
tester. 



TIME- 



K-T-H 



FIG. -. Voltage output of photo- 
tube during operation of iris shutter. 



duration T of the slope is the 
time taken for the edge of the 
shutter opening to pass over the 
slot and is therefore a function of 
slot width. 

As the width of this trapezoid 
near its top is the time recorded to 
indicate speed, it is desirable to keep the slope duration small enough 
to limit the error to 2 per cent. It is therefore necessary that the 
beginning slope and ending slope each be one per cent of the base 
time. This means that the slot width should be one per cent of the 
curtain opening width. 

The minimum curtain opening to be considered is 1 / in., so the 



414 



R. F. REDEMSKE 



Vol 46, No. 5 



929 



slot was fixed at 0.00125 in. For a shutter speed of Viooo sec and a 
one per cent slot, a uniform frequency response up to 100,000 cps 
is required 

/ = i/r = 1/0.001 = 100,000 (2} 

Three phototubes are built 
into the shutter holder jig and, 
as in the iris unit, require long 
cabling to the amplifier, necessi- 
tating cathode followers. The 3 
phototube circuits are similar to 
that shown in Fig. 5, except that 
a 6J5 is used with R L equal to 
150,000 ohms. 

Stylus Control Amplifier. 
For controlling the stylii, the 
circuit of Fig. 7 was used. It is basically a 2-stage direct-coupled 
amplifier using Teledeltos as the output plate load resistor. The 
6SF5 plate load resistor satisfies Eq (2) for 100,000 cps. The 
6SF5 is normally at zero bias and the voltage drop across its plate 
load resistance is sufficient to bias the 6L6 to plate current cutoff . 




FIG. 5. 



Phototube and impedance- 
changer circuit. 




CURTAIN 
SLIT 



SLOTS " 



CURTAIN 




LIGHT 
SOURCES 

MASKS ' 




SHUTTER 



FIG. 6. Optical system of focal-plane shutter tester. 

When a negative signal is applied to the 6SF5 grid, its plate current 
decreases and the negative bias is removed from the 6L6, allowing 
current to flow through the Teledeltos (assuming the solenoid is 
energized so it holds the stylii down on the recording drum). An 
input signal to the 6SF5 grid of one volt is sufficient to produce a 
trace. 



May, 1946 



ELECTRONIC SHUTTER TESTERS 



415 



It was initially planned to ground the recording drum, thus 
grounding the positive 250-v plate supply of the 6L6. However, 
this led to hum difficulties in the 6SF5 grid circuit owing to the float- 
ing power supply circuit. The drum was therefore insulated and 



FRONT PANEL 



6SF5 




FIG. 7. Basic circuit used in each stylus-control amplifier. Plate circuit 
of 6L6 is completed when stylus is pulled down to recording drum by the 
solenoid, because the drum is 250 v positive. 

the ground applied at the negative terminal of the 6SF5 power supply. 
The operator is protected by the mechanical arrangement of the 
drum. For changing recording paper, the drum is hinged outward, 
as shown in Fig. 15, disengaging it from the drive motor and dis- 
connecting it from the high volt- 
age. 

Eleven of these stylus ampli- 
fiers are provided, ten for iris 
shutter recording (three of which 
are used for focal-plane record- 
ing) and one for night photo- 
camera timing, as was shown in 
Fig. 2. All run from a common 
power supply. 

Iris Divider System. The 
first consideration in designing 
the iris divider system is the 
voltage increments on which the 
chain of stylus amplifiers are to 
operate. The first amplifier in 
the chain goes directly to the signal source and will always operate 
on one volt. If the divider is designed so the succeeding channels 



x^ 
/ 
/ 

/ 


-sl i 

X n 9 

=$f 

\6 


> / V. 

\^ RECORDED BASE LINE - J 1 
L--V ACTUAL BASE LINE 
ERROR ERROR 


/^ 


_Siz Q 


/ 


\fl 


/ 


\ 7 


I 


V 


I 


\ ' 


I 


\A 


(B) / 


Y -i 




\ o 3 


^ .^ 


NL-i 


RECORDED BASE LINE' 


ACTUAL BASE LINE , 



FIG. 8. Base-line recording errors 
for 1-v (/I) and 10-v (B) operating 
increments of iris voltage-divider. 



416 



R. F. REDEMSKE 



Vol 46, No. 5 



work on one-volt increments, the situation shown in Fig. 8A will 
exist. This is an undesirable condition as the base line of the record- 
ing is not the true base line. To avoid this difficulty, the divider 
was designed to give 10-v operating increments. This yields a 
recording with negligible error in the base line, as in Fig. 8B. 
The divider resistance was made 100,000 ohms and was built up of 
precision wire-wound resistors having one per cent tolerance, for 
which individual values are given in Table 2. 

TABLE 2 

Voltage Divider Design 



R 1 
R 2 
R 3 
R 4 
R* 
R 
R* 
R 8 
R 
R io 



Per 
Cent 

of 
Total 

90.90 

4.33 
1.54 
0.79 
0.48 
0.32 
0.23 
0.17 
0.14 
1.10 



Value 
in Ohms 

90,900 

4,330 

1,540 

790 

480 

320 

230 

170 

140 

1,100 



Gives Iv 

When 
Input Is 

llv 
21v 
31v 
41v 
51v 
61v 
71v 
81v 
91v 



Amplifier for Iris Shutters. From the preceding discussion, it 
is seen that the amplifier unit for iris shutters must be capable of 
amplifying the phototube output sufficiently to supply the divider 



6SJ7 



6SN7GT 




100 



100 



100 



KX) 5000 10,000 IpOO 
B+ 



FIG. 9. Amplifier circuit used between phototube and iris voltage-divider 
when testing iris shutters. 

with a 91-v pulse without appreciable distortion. The components 
were selected by using Eq (1) on the basis of passing up to 10,000 
cps. The resultant circuit is shown in Fig. 9. Gain control RI is 



COUPLING 



May, 1946 ELECTRONIC SHUTTER TESTERS 417 

necessary to compensate for the difference between maximum open- 
ing areas of the various shutter types. A position is selected for each 
shutter type so that the amplifier output to the divider at full shutter 
opening is the 91 v required so that the last stylus amplifier just 
operates to form a trace. 

The smallest shutter gives a % 2-v signal at full opening, so the gain 
is based on this figure, and all larger shutters are scaled down to this 
value by RI. 

Potentiometer R 2 is provided to oppose the steady direct voltage 
present across the cathode follower resistance, so that no d-c potential 
exists across the gain control. In addition, a 100-ohm potentiom- 
eter adjusts the bias on the 6SJ7. 

With the phototube totally dark and the system warmed up, R z 
is adjusted until moving RI slowly from one end to another does not 
cause the plate current meter 
to vary. The bias is then ad- 
justed to the correct value as in- 
dicated by zero indication of the 
plate milliammeter. Controls RI 
and R 2 are screwdriver adjust- 
ments and only have to be re- 
aligned occasionally as a routine FIG. 10. Overshoot distortion 
check caused by use of isolating capacitor C 

in Fig. 9. 

When the system is to be set 

up for a new-type shutter, the shutter is opened to its wide-open 
position and is illuminated. This gives the maximum value of light 
that the phototube will receive when the shutter is operated nor- 
mally. Control RI is adjusted to give a reading of 1.2 ma on the 
meter, thus insuring that the top of the phototube output pulse will 
just be the top of the recorded pulse. The cathode follower is used 
as before to act as an impedance changer and minimize shunt capaci- 
tance effects. 

It is theoretically desirable to keep the system a d-c amplifier 
throughout, to handle the d-c component that exists in trapezoidal 
pulse. However, the practical difficulties encountered in the system 
dictated the use of an isolating capacitor C. Because the capacitor 
will not pass direct current, there is some distortion of the pulse as in 
Fig. 10, but by suitable choice of RC ratio the effect can be reduced 
to the point where it is not serious. 

Amplifier for Focal-Plane Shutters. The function of the amplifier 




TIME 



418 



R. F. REDEMSKE 



Vol 46, No. 5 



for focal-plane shutters is to raise the level of the optical system 
output to the voltage necessary to excite the stylus control to form 
a trace. The maximum output of the phototube is one millivolt 
and the voltage required to draw a trace is one volt, so the require- 
ment is an amplifier with a gain o'f 1000 or more and a frequency 



6SJ7 



G5 



0.25 




OUTPUT 



FIG. 11. Amplifier circuit used between phototube and stylus-control 
amplifier when testing focal-plane shutters. 

response to 100,000 cps. The circuit of Fig. 11 is the result of de- 
signing to these figures. As the lowest shutter speed is Vi25 sec and 
as no steady-state light calibration is needed, the amplifier is capaci- 
tance-coupled throughout. The cathode-follower output is used to 
provide a low-impedance line to the stylus amplifier. Three of these 
amplifiers are provided, one for each phototube output. It was 
originally planned to use type 
931 photomultipliers for the 
focal-plane tester because of 
the low light level encountered, 
but these tubes were critical at 
the time of this development 



KI9B PHOTOTUBE 



0.2 FOOT-CANDLE 
NEUTRAL FILTER 




TIMING/ \ 
PHOTOTUBE 1 ' 



SHUTTER 



FIG. 12. Optical system of night 
photo shutter tester. 



and could not be obtained. 
Theu: use would have eliminated 
the need for the high-gain 
amplifier. 

Night Photo Timing. The 
K-19 night photo unit required an auxiliary circuit for adequately 
testing it. This camera has a magnetically actuated iris-type 
shutter which is controlled by a phototube-amplifier arrangement. 
In use, the light from a flash bomb dropped from the camera 
airplane reaches the phototube, actuating the shutter through an 
amplifier. The requirement is that the shutter shall be fully open in 



May, 1946 ELECTRONIC SHUTTER TESTERS 419 

10 milliseconds from the start of the flash, at which time the flash has 
reached peak intensity. The system must work on a light intensity 
change of 0.2 ft-c. 

In the testing of this camera, the shutter is tripped by a pulse of 
light which simultaneously actuates a phototube in the tester. The 
light pulse is produced by operating a shutter located in front of the 
light source, as shown in Fig. 12. The shutter speed is 1 / M o sec, 
which gives a light pulse wave front similar to the actual flash bomb. 
This is important because the K-19 camera operates on rate of 
change of light rather than steady-state values. This light pulse 
passes through the shutter and strikes an optical dividing system con- 
sistihg of a piece of glass set at 45 deg to the light path. 



U 10 MILLISECONDS -*j 

[ | FULL SHUTTER OPENING 

LIGHT/ \TIMING PULSE S* V 

APPLIED FROM NIGHT / \ 

PHOTO CHANNEL / ~\ 



FIG. 13. Timing record obtained when checking night photo shutters. 

Most of the light passes straight through and strikes the cathode 
of the timing phototube, while a small portion is reflected by the 
glass to actuate the phototube of the K-19 camera and trip its 
shutter. This reflected light is further attenuated by a neutral filter 
so the maximum light reaching the K-19 phototube is 0.2 ft-c. The 
light reaching the timing-channel phototube is amplified and fed 
to the stylus control to form a line. The beginning of this line is 
the exact instant that the K-19 phototube was energized, so the 
distance to full open shutter can be measured to see if it is within the 
10 milliseconds allowed. A typical record is shown in Fig. 13. 

Direct-Indicating Shutter Tester. The direct-indicating tester 
was developed as a general utility instrument to be used separately 
or in conjunction with the recording tester. To obtain a high 
degree of accuracy from the indicator, the output reading is pre- 
sented as per cent deviation from rated shutter speed. In this in- 



420 



R. F. REDEMSKE 



Vol 46, No. 5 



strument, iris shutter speed is defined as the total open time, which 
is the time duration of the base of the characteristic trapezoid. 

The final design of this unit consists of a time-measuring circuit, 
a comparing circuit, and an indicating circuit, as in Fig. 14. 

Time-Measuring Circuit. The time-measuring circuit makes use 
of the constant-current characteristic of a pentode to charge a 
capacitor to a voltage proportional to time. The input is received 
from the amplifier output of the iris or focal-plane recording testers 
previously described, or from a similar circuit when used separately. 



INPUT 




FIG. 14. Circuit of direct-indicating shutter tester. 



The 6SF5 tube acts as a combination amplifier and limiter. It is 
run normally at some bias such that the drop through its plate load 
resistor keeps the 6SJ7 biased beyond plate current cutoff. A small 
negative signal then will drive the 6SF5 to cutoff and the 6SJ7 will 
become conductive and start charging the capacitor in its plate cir- 
cuit at constant current. This charging current flowing in the 
cathode circuit and in the divider causes some regeneration owing 
to RI and Rz, which effectively increases the gain of the circuit but is 
not sufficient to cause instability. The effect is somewhat counter- 



May, 1946 



ELECTRONIC SHUTTER TESTERS 



421 



acted by the degeneration in R 3 , which is provided to keep the 
6SJ7 grid in a region of greater current linearity. 

For iris shutters the arrangement is such that the 6SF5 is driven 
to cutoff by a shutter opening area equal to one per cent of total 
opening, remaining in that condition until the shutter is within one 
per cent of being closed. Thus the capacitor is being charged at 
constant current for the full open time of the shutter. With focal- 
plane shutters, the trapezoidal pulse described previously is of suf- 
ficient amplitude to drive the 6SF5 to cutoff and charge the capacitor 
for its duration. 

One capacitor (d, C 2 , or 3) is used for each shutter speed to be 
covered, ;and a switch is pro- 
vided to select the proper one. 
The capacitor values are so 
chosen that each one when 
charged for its indicated shutter 
speed will develop the same 
voltage. 

Comparator Circuit. Switch 
Si is a telephone-type switch 
that connects the 6SJ7 plate cir- 
cuit to the capacitor selected by 
the speed selector switch. When 
the shutter is operated and the 
capacitor charged, Si is thrown 
to its other position* This dis- 
connects the capacitor from the 
plate circuit and places it in 
series opposition with a standard 
source of voltage. 

If the shutter speed is exact, 




FIG. 15. Recording drum of record- 
ing shutter tester, shown hinged out- 
ward for changing of paper. Record- 
ings of 4 shutters can be obtained on 
a single sheet of Teledeltos paper before 
reloading. 



the 2 voltages will be equal and 
their resultant will be zero. If the shutter is in error, the voltages 
will differ and the resultant will indicate the direction and amount 
of error. 

Switch Ss is provided to completely discharge the capacitor before 
repeating the charging process. 

It was found impossible to conveniently select capacitor values 
close enough to give an exact voltage for a given charging time, so 
voltage trimmers R^, R&, and R 6 were added. One of these is selected 



422 R. F. REDEMSKE Vol 46, No. 5 

with each capacitor value and is adjusted to give the proper compari- 
son voltage. This circuit is calibrated by applying electrical pulses 
of known time duration from a commutator arrangement to the 
amplifier input. 

Indicator Circuit. The resultant of the capacitor and comparison 
voltages is fed into an indicator circuit through a 30-megohm re- 
sistor. This value is made high so the capacitor will discharge 
slowly enough to provide a reliable reading for conditions of shutter 
error. 

The 6J5 circuit is essentially a vacuum-tube voltmeter with a zero- 
center indicator. Shutter speeds higher than rated result in an up- 




FIG. 16. Direct-indicating shutter tester, with fixture for phototube, light 
source, and shutter holder at the right. 



ward deflection, low speeds in a downward deflection. As the volt- 
ages being compared for any shutter speed are the same, the indicator 
can be calibrated in percentage deviation from rated speed. The 
slight variations owing to R 4> R$, and RQ do not introduce appreci- 
able error. The full meter scale covers = fc 15 per cent error. 

The circuit constants are such that large inputs cannot damage 
the 50-050 pa, meter, as the 6J5 reaches saturation slightly off scale 
on the upper end and reaches cutoff beyond the lower end. 

Several of these testers were built, for various numbers of shutter 
speeds. The unit shown in the circuit was arranged for shutter 
speeds of l / m , l /w, and l / m sec. 

The indicating tester proved an accurate and convenient means of 



May, 1946 ELECTRONIC SHUTTER TESTERS 423 

measuring shutter speed. On some production runs it was used 
exclusively for speed tests, the recording instrument being resorted 
to only for occasional spot checks. It was also useful in production 
adjustment of shutters. 

The author is indebted to Irving Doyle, Fairchild engineer, and 
Robert Nelson of the inspection department for substantial coopera- 
tion and many helpful ideas in connection with the development of 
the recording shutter tester. 



BOOK REVIEWS 

Television: The Eyes of Tomorrow. By William C. Eddy. Prentice-Hall, 
Inc. (New York), 1945, 330 pp.; illust. and diagrams; 6X9 in., $3.75. 

This book is written by a man who has had much experience in the television 
field. It is strongest and most valuable where it makes available to others the 
personal experiences of the author. His selection of secondary material, that is, 
the experiences of others, also adds to the value of the book as a record of prac- 
tical experience. 

The chapter "The Use of Film in Television" outlines experience with a variety 
of types of film presentation and ends with the conclusion : 

"Whether or not film in television will increase in importance in future 
commercial programming remains an economic rather than an engineering 
problem. Based on the present development in this field, coupled with the 
possibilities that are known to exist in future improvements of the system, 
the popularity of film programs in the home is solely dependent on the enter- 
tainment value of the product. It is hoped that a satisfactory solution to the 
present economic impasse between broadcasters and film distributors will 
someday be effected and that eventually television will have a satisfactory 
film product available for programming." 

The book is weakest in the chapters dealing with the technical aspects of 
television, as technical accuracy has been sacrificed at times in the attempt to 
secure a popular style of presentation. This is probably relatively unimportant 
in view of the main audience to which the book is addressed. The author was 
well aware of the hazards involved in writing a book of this type and disarms 
one by his remarks in the preface: 

"Any person who deliberately takes it upon himself to write a book on 
television asks for criticism of both qualified critics and lay readers. . . . 

"Experts in each field will necessarily point out wide discrepancies between 
my coverage of their specialty and the practical application of their trade on 
the studio floor." 

D. R. WHITE 
Mar. 20, 1946 

Television Show Business. By Judy Dupuy. General Electric Company 
(Schenectady, New York), 1945, 246 pp.; illust. and diagrams; 8 l / 2 X 11 in., 
$2.50. 

This handbook of television programming and production, based on 5 years 
of operation of General Electric's Television Station, WRGB, Schenectady, 
New York, is the most complete and detailed account which has yet appeared 
in print of experience with television programs. As such it will be an invaluable 
guide to others in the television field. The treatment of the technical aspects of 
television is reduced to a brief popular account of little interest to an engineer 
but adequate for the intended reader. 
424 



BOOK REVIEWS 425 

The chapter "Films in Television" is probably the one of greatest interest to 
the Society members. This outlines various functions performed by films of 
different types, including their use as a supplement to live talent shows in order 
to circumvent certain of their limitations. 

In common with other books on television this one involves a certain amount 
of prophecy. Concerning the role of motion pictures in television it says: 

"Films will play a major role in television. It is safe to predict that future 
television programming will consist of about one-third studio shows, one- 
third remote pickups (mobile unit or distant permanent setups), and one- third 
motion pictures. These motion pictures may or may not be made exclusively 
for television. At first many of them will be motion pictures produced for 
movie theater showing. 

"The future holds many unknown factors which will determine the relation- 
ship of motion pictures and television." 

The General Electric Company has made a very worth-while contribution 
to the television industry in thus making available its experience at WRGB. 

D. R. WHITE 
Mar. 20, 1946 

Report of Conference on Unification of Engineering Standards, Ottawa 
Canada, September-October, 1945. Combined Production and Resources 
Board. 90 pp.; 6 X 9 in. Superintendent of Documents, U. S. Government 
Printing Office, Washington 25, D. C.; 20 cents. 

This report summarizes the proceedings of the third conference of representa- 
tives of the Canadian, British, and American standards associations on the 
question of unification of standards for screw threads, limits and fits, drawing 
practice, and metrology in countries using the "inch" system of measurement. 
Final agreement was reached on unified standards for acme and acme stub 
screw threads, buttress screw threads, fastening screws of the smaller sizes, and 
fine motion screws. 

The possibility of developing a common standard for screws for camera and 
lens attachments, and the like, and for tools for screw thread production was 
referred to committees of the 3 standards associations for further consideration. 
A proposed specification for a basic screw thread form was prepared for sub- 
mission to the industries concerned for comments. 

No agreement was reached on unified standards for high duty studs in light 
alloys or for pipe threads. Two alterations in American War Standard B1.6 
(June, 1945) on truncated Whitworth threads were recommended and referred 
to the American Standards Association for study. 

Included in the report is a bibliography of standards to which reference was 
made during the conference, and tables listing the proposed screw thread series 
and screw thread symbols. 

M. WRIGHT 
Mar. 7, 1946 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 



Acoustical Society of America, Journal 

17, 4 (Apr., 1946) 
Stereophon Sound Recording System (p. 356) H. B. LEE, III 

American Cinematographer 

27, 2 (Feb., 1946) 

The Subjective Camera (p. 46) H. A. LIGHTMAN 

Report of the SMPE Committee on Color (p. 48) 

Bell & Howell Introducing New 16-Mm Auto Load 
Camera (p. 52) 

Bell & Howell Equipment to Be Manufactured in 
England (p. 54) 

Eastman Kodak Explains Research Status of Pro- 
fessional Safety Film (p. 54) 

New Western Electric Sound Recorder for 16-Mm 
and 35-Mm (p. 60) 

27, 3 (Mar., 1946) 

Cinematography in the War Filming Projectile Tests 

for the Navy (p. 80) A. L. GILKS 

The Fluid Camera (p. 82) H. A. LIGHTMAN 

A % Positive Vari-Focal View Finder for Motion Picture 

Cameras (p. 84) F. G. BACK 

Biological Photographic Association, Journal 

14, 3 (Mar., 1946) 
High-Speed Motion Picture Photography (p. 107) H. M. LESTER 

British Kinematograph Society, Journal 

8, 4 (Oct.-Dec., 1945) 
Motion Picture Laboratory Trends and Practices 

(p. 73) W. M. HARCOURT 

Electronics 

19, 4 (Apr., 1946) 

Color Television on Ultra High Frequencies (p. 109) 
426 



CURRENT LITERATURE 



427 



Electronic Industries 

5, 3 (Mar., 1946) 

CBS Shows Its Color High-Frequency Transmission 
with Sound on Same Carrier Demonstrates Their 
Solution of Color Reception Problems (p. 75) 

RCA Color TV Status Princeton Laboratories 
Demonstration Reveals Progress in Both Color and 
Black-and-White Tube and Transmitter Produc- 
tion (p. 102) 

International Photographer 

18, 1 (Feb., 1946) 
Device Makes Dissolves and Fades with Camera 

(P- 7) 
Practical Utilization of Monopack Film (p. 11) 

18, 2 (Mar., 1946) 

Processing 16-Mm Ansco Color Film (p. 5) 
Color Television When? (p. 20) 

Some Considerations in Using Kodachrome, Pt. 2 
(P. 22) 

International Projectionist 

21, 2 (Feb., 1946) . 

Some Physical Properties of Film Relating to Image 
Stability (p. 7) 

The New Ampro 16-Mm Projector, Pt. 2 Step-by- 
Step Analysis of Sound Circuit (p. 12) 

Acetate vs. Nitrate Issue Revived (p. 15) 

Basic Design Determines Projector Performance 
(P- 16) 

Projection Lens Aberrations (p. 20) 

Basic Radio and Television Course, Pt. 20 Trans- 
mitters (p. 22) 

21, 3 (Mar., 1946) 

Elements of Projection Optics (p. 7) 

An Improved Loudspeaker System (p. 10) 

Basic Radio and Television Course, Pt. 21 Trans- 
mitters (Oscillators) (p. 18) * 

Proceedings of the I.R.E. 

34, 2 (Feb., 1946) 

Transmission of Television Sound on the Picture 
Carrier (p. 49) 



Technique Cinematographique, La 

17, 15 (Feb. 5, 1946) 
A New Micro-Densitometer (p. 275) 



J. YOLO 

C. G. CLARKE 

J. L. FORREST 
W. S. STEWART 



R. H. TALBOT 
L. CHADS OURNE 



A. E. MEYER 
A. MONTANI 

M. BERINSKY 

A. MONTANI 
J. B. LANSING AND 
J. K. MILLIARD 

M. BERINSKY 



G. L. F&EDENDALL, 

K. SCHLESINGER AND 
A. C. SCHROEDER 



L. RAITIERE 



SOCIETY ANNOUNCEMENTS 



ATLANTIC COAST SECTION MEETING 

One of the largest meetings of the Atlantic Coast Section of the Society was 
held on March 13 at the new DuMont John Wanamaker Studios when Dr. Allen 
B. DuMont, president of A. B. DuMont Laboratories, Inc., delivered a paper 
on "The Relationship of Television to Motion Pictures." The linking of tele- 
vision and motion pictures in a partnership of mutual advantages, with the 
possibility of motion picture producers soon making films expressly for television, 
was seen by Dr. DuMont. 

Among the advantages cited were that films provide a permanent record for 
use any time and place, film programs can be handled with a minimum of tech- 
nical personnel in the studio, and it provides an effective means of syndicating 
production among any number of stations. 

"Movies and television are natural partners," Dr. DuMont said. "One supple- 
ments the other. Movies are the permanent record. Television is the more 
advanced way of getting the picture. Television owes much to movies up to this 
time. But from here on, movies will be receiving benefits from the rapidly refining 
television technique . ' ' 

The meeting concluded with a demonstration of a sight and sound television 
motion picture film. 

The last meeting in the Spring series arranged by the SMPE Section to promote 
wider knowledge of industry techniques and practices was devoted to the produc- 
tion of newsreels, under the title "The Newsreel Its Production and Signifi- 
cance." Held at Movietonews Studios, on April 17, the symposium consisted of 
Dan Doherty, assignment editor, Harry Lawrenson, foreign editor, Warren 
McGrath, sound engineer, Jack Gordon, unit director, Vyvyan Donner, women's 
editor, Walter Mclnnis, cameraman, and Bert Hoist, librarian. 

Mr. Doherty described the status of the newsreel as a screen journal. He ex- 
plained why the reel does not attempt coverage like newspapers. "The newsreel 
editor cannot, for many reasons, attempt to cover all the news. In the first place 
the cost of keeping a camera staff capable of the noble effort would be prohibitive. 
In the second place, newsreel presentation time in theaters is limited. 

"By this confining fact alone, the editor's task is not one of attaining total 
coverage but of selection, and selection based on an intimate knowledge and 
understanding of the medium." 

The Movietone editor defended the practice of presenting the same annual 
events football games, horse races, etc. on the grounds that audiences wanted 
them. Mr. Doherty added that other trivia, such as cheesecake and fashion 
shows, were the result of policy which required an entertaining newsreel. 

Mr. Doherty introduced each of the other newsreel experts who discussed their 
special work in newsreel and fashion short production. 
428 



SOCIETY ANNOUNCEMENTS 429 

These papers, as well as the paper by Dr. DuMont, will be published in a forth- 
coming issue of the JOURNAL. 

PACIFIC COAST SECTION MEETING 

The March 20 meeting of the Pacific Coast Section of the Society was addressed 
by Randal Terraneau of the Humphries Laboratory, London, who is currently 
making his first visit to Hollywood in several years. Mr. Terraneau 's witty de- 
scriptions of laboratory operations under blitz conditions in London were highly 
interesting. 

Other speakers at the meeting were W. C. Kunzman, L. L. Ryder, E. A. Willi- 
ford, and Carrol Dunning, who introduced Mr. Terraneau. Approximately 150 
members and guests were present at the discussion, held at the Hollywood 
Athletic Club. 

On March 26 the Section was privileged to hear talks by Dr. H. F. Olson and 
Dr. V. K. Zworykin of the RCA Laboratories, Princeton, New Jersey. Dr. Olson 
described the Acoustic Laboratory at Princeton, a reverberation simulator, a 
duo-cone loudspeaker, and a miniature unidirectional microphone. 

Dr. Zworykin discussed the status of television including the Image Orthicon 
and the Aluminized Kinescope. He also reviewed the advancements in the 
electron microscope, and described a small electron viewing tube suitable for use 
under invisible infrared light conditions. 

A record-breaking attendance of approximately 600 members and guests 
completely filled the Walt Disney Studio theater at Burbank, Calif. 



EMPLOYMENT SERVICE 

Position available for Optical Designer, capable of handling the calcula- 
tion and correction of aberrations in photographic and projection lens 
systems. Junior designers or engineers will be considered. Write 
fully giving education, experience, and other qualifications to Director 
of Personnel, Bell and Ho well Company, 7100 McCormick Road, Chi- 
cago 45, 111. 



Motion picture studio in Bombay, India, has positions open for profes- 
sional motion picture camerman with studio and location experience; 
sound recording engineer experienced in installation, maintenance and 
operation of recording equipment; motion picture processing labora- 
tory supervisor; and professional make-up artist. Five-year contracts 
at favorable terms are offered to those qualified. Write or cable direct 
to Personnel Manager, Dawlat Corporation Ltd., Patel Chambers, French 
Bridge, Bombay 7, India, giving experience, etc., in detail. 



New film production unit to be located at Athens, Georgia, needs film 
editor-writer and film director. Experience in 16-mm as well as 35-mm 
production desirable. Southern background or interest in South pre- 
ferred but not essential. Write giving full details of experience, etc., to 
Nicholas Read, The National Film Board, Ottawa, Canada. 



430 SOCIETY ANNOUNCEMENTS 

POSITIONS WANTED 

Projectionist-newsreel editor with 15 years' experience just released 
from service. Willing to locate anywhere. Write P. O. Box 152, Hamp- 
. den Station, Baltimore 11, Maryland. 



Honorably discharged veteran with 10 years' experience in projection 
and installation of projection and sound equipment, both for booth and 
back-stage. Prefer to locate in California, Oregon or Nevada. For ad- 
ditional details write F.A.N., Box 113, Holley, Oregon. 



Cameraman, honorably discharged Army veteran, desires re-enter indus- 
trial, educational production with independent producer or studio. Ex- 
perienced in 35- and 16-mm color and black-and-white. References 
and complete record of experience available. Write, wire or telephone 
T. J. Maloney, 406 Oak St., Ishpeming, Mich. Telephone 930. 



We are grieved to announce the death of Robert M. Johnston, Asso- 
ciate member of the Society, on September 19, 1945, in Rock Falls, 
Illinois. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 JUNE, 1946 No. 6 

CONTENTS 

PAGE 

An Appraisal of Illuminants for Television Studio 
Lighting R. E. FARNHAM 431 

Carbon Arcs for Motion Picture and Television Studio 
Lighting 

F. T. BOWDITCH, M. R. NULL AND R. J. ZAVESKY 441 

An Improved Film-Drive Filter Mechanism 

C. C. DAVIS 454 

A Simplified All-Purpose Film Recording Machine 

G. R. CRANE AND H. A. MANLEY 465 

The Use of Desiccants with Undeveloped Photographic 
Film C. J. KUNZ AND C. E. IVES 475 

American Standard 16-Mm Test Films 511 

Standardization and the Antitrust Laws J. D. HAYES 516 

Current Literature 526 

Society Announcements 527 

Program of the Fifty-Ninth Semi-Annual Technical 
Conference 528 

Index to Journal, Vol 46 (January- June, 1946) : 

Author Index 536 

Classified Index 539 



Copyrighted, 1946, by the Society of Motion Picture Engineers, Inc. Permission to republish 
material from the JOURNAL must be obtained in writing from the General Office of the Society. 
The Society is not responsible for statements of authors or contributors. 

Indexes to the semi-annual volumes of the JOURNAL are published in the June and December 
issues. The contents are also indexed in the Industrial Arts Index available in public libraries. 



JOURNAL 

OF THE 

SOCItry of MOTION PICTURE ENGINEERS 



UOTCL PENNSYLVANIA NCW YOftKl, N-Y- 



TCL. PCNN. 6 O62O 



HARRY SMITH, JR., EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CHARLES W. HANDLEY 

ARTHUR C. HARDY 

Officers of the Society 
^President: DONALD E. HYNDMAN, 

350 Madison Ave., New York 17. 
*Past-P resident: HERBERT GRIFFIN, 

133 E. Santa Anita Ave., Burbank, Calif. 
*Executive Vice-P resident: LOREN L. RYDER, 

5451 Marathon St., Hollywood 38. 
** Engineering Vice-P resident: JOHN A. MAURER, 

37-01 3 1st St., Long Island City 1, N. Y. 
*Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland 1, Ohio. 
**Financial Vice-President: M. R. BOYER, 

350 Fifth Ave., New York 1. 
* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland 1, Ohio. 
* Secretary: CLYDE R. KEITH, 

233 Broadway, New York 7. 
^Treasurer: EARL I. SPONABLE, 

460 West 54th St., New York 19. 

Governors 

"fFRANK E. CAHILL, JR., 321 West 44th St., New York 18. 
**FRANK E. CARLSON, Nela Park, Cleveland 12, Ohio. 
**ALAN W. COOK, Binghamton, N. Y. 

*JOHN I. CRABTREE, Kodak Park, Rochester 4, N. Y. 

*CHARLES R. DAILY, 5451 Marathon St., Hollywood 38. 
**JOHN G. FRAYNE, 6601 Romaine St., Hollywood 38. 
**PAUL J. LARSEN, 1401 Sheridan St., Washington 11, D. C. 
**WESLEY C. MILLER, Culver City, Calif. 

*PETER MOLE, 941 N. Sycamore Ave., Hollywood. 
"JHoLLis W. MOYSE, 6656 Santa Monica Blvd., Hollywood. 

*WILLIAM A. MUELLER, 4000 W. Olive Ave., Burbank, Calif. 
*A. SHAPIRO, 2836 N. Western Ave., Chicago 18, 111. 

*REEVE O. STROCK, 111 Eighth Ave., New York 11. 

*Term expires December 31, 1946. tChairman, Atlantic Coast Section. 
**Term expires December 31, 1947. jChairman, Pacific Coast Section. 
* Chairman, Midwest Section. 



Subscription to nonmembers, $8.00 per annum; to members, $5.00 per annum, included in 
their annual membership dues; single copies, $1.00. A discount on subscription or single copies 
of 15 per cent is allowed to accredited agencies. Order from the Society at address above. 
Published monthly at Easton, Pa., by the Society of Motion Picture Engineers, Inc. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York 1, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton. 

Pa., under the Act of March 3. 1879. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

Vol 46 JUNE, 1946 No. 6 



AN APPRAISAL OF ILLUMINANTS FOR TELEVISION STUDIO 

LIGHTING* 

R. E. FARNHAM** 

Summary. The paper first outlines the 'requirements of an illuminant for tele- 
vision studio lighting, including data as to the illumination levels as well as the quality 
of the light necessary. Then follows a discussion of each of several possible sources, 
such as mercury vapor lamps, tungsten filament types, and fluorescent tubes. An 
appendix is included, giving tabular data as to the energy available in the ultraviolet, 
visible, and infrared regions. 

Thus, the paper makes possible an accurate judgment of the advantages and dis- 
advantages of several illuminants adapted to this application. 

In making an analysis of the various light sources applicable for 
television studio illumination, it is necessary to make certain general 
assumptions regarding the characteristics of the several elements in 
the light path. These are as follows : 

(1) Spectral Sensitivity of the Iconoscope. The light pickup 
surface of the iconoscope generally employs a caesium or modified 
caesium coating whose sensitivity to energy radiation is best illus- 
trated in Fig. 1. It is interesting to note that the iconoscope 
sensitivity is practically the reverse of that of the eye, which has its 
peak in the yellow-green region (5500 A), thus color values will be 
badly distorted unless corrective means are taken. Deep reds and 
violets, ordinarily considered as the darker colors, tend to appear 
bright, while yellow and green go dark. 

(2) Requisite Illumination Levels. Based on tungsten illumina- 
tion, approximately 1000 to 1200 ft-c, vertical surface illumination, 
are generally required. This presumes the television camera lens 
to be operating at its maximum satisfactory aperture (/4.5). 

* Presented Sept. 27, 1944, at a meeting of the Atlantic Coast Section of the 
SMPE in New York. 

** General Electric Company, Nela Park, Cleveland, Ohio. 

431 



432 



R. E. FARNHAM 



Vol 46, No. 6 



(3) Reflection Characteristics of Televised Areas. Inasmuch as 
the quality of the television image is judged largely by the appear- 
ance of the faces, skin reflection characteristics constitute an im- 
portant element in the effectiveness of a light source. 

It is evident from Fig. 2 that in spite of a high iconoscope sensitivity 
in the violet and near-ultraviolet regions, an illuminant rich in violet 
and ultraviolet energy is going to be greatly reduced in its effective- 
ness when reflected from the faces of the actors. Dark clothes and 
properties may oftentimes be good reflectors of violet and ultraviolet 
energy with the result that faces may appear darker by comparison. 



500 




1000 2000 3000 4000 5000 6000 7000 8000 9000 IOOOO 
WAVELENGTH - Angstrom Units 

FIG. 1. Spectral distribution of photoelectric surface. 



On the other hand, the combination of the iconoscope's 7500 A 
sensitivity peak, the high reflectivity of skin, and the strong energy 
radiation from tungsten at 7500 A tends to accentuate the faces in the 
television image to the point of loss of detail. In other words, faces 
become chalky white when the attempt is made to show detail in the 
darker parts of the setting. 

Other general factors affecting the choice of television illuminants 
are as follows: 

Heat on the Set. Experience gained in connection with motion 
picture studio lighting indicates that at 100-200 ft-c illumination 
from unfiltered tungsten sources, a noticeable though not uncom- 
fortable warmth is felt. At 400-500 ft-c, the infrared radiation 



June, 1946 



APPRAISAL OF ILLUMINANTS 



433 



produces a definite discomfort, particularly in prolonged scenes. 
At the 1000-1200-ft-c level necessary for television, the discomfort 
is immediate and means must be taken to reduce the volume of 
infrared radiation such as filtering or the set cooled by positive 
ventilation. 

Electrical Interference. The light output of electric discharge 
sources, such as mercury and some fluorescent lamps, follows the 
cyclic variation of the a-c supply quite closely. This 120-cycle 
light ripple has not in general been difficult to neutralize. The 
neutralization problem can be somewhat simplified by operation of 
these sources in groups of three, from a 3-phase supply, greatly re- 



300 400 

WAVELENGTH - Millimicrons 



FIG. 2. Reflection factor of skin (upper forearm) . 



ducing the amount of ripple and increasing its frequency to 360 
cycles. 

Radio frequency impulses are sometimes produced in circuits 
supplying electric discharge lamps. These arise from a behavior of 
the arc stream or in the auxiliaries. Their elimination is completely 
and easily accomplished by suitable filters. 

Operation on d-c is in many cases entirely practicable, in the event 
that a-c operation presents extreme difficulty. 

Sources. The following commercially available illuminants have 
been considered: 

Mercury 

1000-w AH-6 water-cooled lamp 

3000-w AH-9 lamp 
Tungsten filament lamps 



434 R. E. FARNHAM Vol 46, No. 6 

Motion picture studio types 
Reflector bulb types 
Fluorescent lamps 

Mercury Lamps. The merits of the AH-6 water-cooled lamp have 
given it a predominant position as a television light source. Its 65,000 
lumen output makes it possible to obtain adequate illumination with 
relatively few lighting units. The water cooling necessary in the 
operation of the lamp possesses the additional feature of removing the 
greater part of the infrared or heat radiation, making it by far the 
most comfortable illuminant on the basis of equal foot-candles of 



HOTOELECTRIC SURFACE 
TION FROM SKIN 
a ON 5 

B 8 8 8 














i 






















































Water cooled type H-6 mercury 






**rnn 






I 














^''~ 


Tungsten at 3360 K 
-v^thru 2.3 cm of water 


Eg 

2 

KUJ 40O 




1 












TT 


E/ 




\ 




o 1 *- 
u" 
>(r 200 




M 


1 




I 






UL 


,~r~ 


- _j 






gtf 
1 


_J 


_. 


^ 


rr 


-''' 


-~^~ 




^^ 


,^x 



3000 4000 5000 6000 7OOO 8000 9000 10000 

WAVELENGTH - Angstrom Units 

FIG. 3. Theoretical iconoscope output, making allowance for skin reflection 

characteristics. 



any of those considered. The heat from 1000 ft-c of AH-6 light is 
hardly apparent. 

Operating at an internal pressure of approximately 110 atmos- 
pheres, this source contains considerably more red in its spectrum 
than other mercury sources, taking some advantage of the caesium 
cell's peak at 7500 A and the high reflectivity of skin in this region. 
At the same time, the violets, blues, and even greens and yellows are 
adequately reproduced. These characteristics can best be suited by 
reference to the solid curve of Fig. 3, which gives the theoretical 
iconoscope output, making allowance for skin reflection characteris- 
tics. 

The rated life of the AH-6 lamp is 75 hr (for half -hour burning 
periods) with considerable spread in the life of individual lamps. This 



June, 1946 APPRAISAL OF ILLUMINANTS 435 

may make necessary fairly frequent removals in a large installation 
operated for long periods. The multiplicity of water connections, 
particularly overhead, with the possibility of leakage, might be con- 
sidered a further disadvantage. However, the water pressures em- 
ployed are not high and experience has shown that, with frequent 
inspection, troubles from this cause are few. 

A very complete discussion on the use of the AH-6 lamp for tele- 
vision studio lighting has been given by Breeding. 1 The paper de- 
scribes 3-lamp equipment adapted to remote control of light direction, 
as well as an installation of these units at television station WRGB 
at Schenectady. Operating data are also given. 

The 3000- w AH-9 mercury lamp has been considered as a source for 
television studio lighting, and a limited amount of test data have 
shown its practicability, particularly for lighting large areas. Its 
chief advantages are high light output, 120,000 lumens, and 2000 hr 
life. The spectral energy distribution of the AH-9 is similar to that 
of the AH-6, except that there is a negligible amount of energy emitted 
beyond 6000 A. Thus that part of the solid line curve of Fig. 3 to the 
left of the 6000 A point may be used in analyzing the results obtained 
with the A H-9 lamps. * Faces appear somewhat darker. Possibly the 
AH-9 lamp might be used in combination with tungsten filament 
sources to obtain a better balance in the lighting results. 

The AH-9 emits approximately 13 per cent of its energy in the re- 
gion beyond 7400 A as compared to less than half that from the 
AH-6 lamp, making it noticeably warmer to work under. With 
the addition of sufficient unfiltered tungsten light to obtain a satis- 
factory image appearance, working conditions might approach definite 
discomfort. 

Where 3-phase operation is necessary and three AH-9's are grouped 
closely together, the total of 360,000 lumens may be greater than is 
desirable from a single unit, particularly if mounting heights are 
limited. 

There are no fixtures available for the AH-9 lamp specifically 
designed for television lighting service; however, the large trough 
reflectors developed for high bay lighting applications of the AH-9 
should be satisfactory with modification of the suspension arrange- 
ments. 

* The 2 curves of Fig. 3 are based on 1000-w lamps. When applying the mer- 
cury curve to the A H-9 lamp, the relative output values should be increased 50 
per cent because of the lamp's greater wattage (and lower efficiency). 



436 



R. E. FARNHAM 



Vol 46, No. 6 



Filament Lamps. There are 2 general types of incandes- 
cent lamps applicable to television lighting the high-wattage group 
of 1-, 2-, 5-, and 10-kw rating developed primarily for motion picture 
studio lighting, and the so-called "reflector bulb" types currently 
available in 150- and 300-w sizes. 

Earlier television studio techniques were patterned after those of 
the motion picture studios, the television camera being substituted 
for the motion picture camera. High-wattage "spots" and "floods" 
similar to those of Hollywood were employed. 

However, the necessity of changing smoothly from one setting to 
the next as the television program progresses made the use of these 



RELATIVE INTENSITY 

M*o><6f\}5o 

OOOOOOOOC 
























































































COLOR COORDINATES 
LCI. SYSTEM (1931) 
X=.3I3 Y=.323 


























































































/ 


s"^ 


~^~~^ 


^ 






r 


*V 


^ 


























\, 










\ 














- 














"^ 






s 




































\ 












/ 






















\ 




































\ 








/ 


























\ 




































\ 




' 
































Nj 





































WAVELENGTH IN ANGSTROMS 

FIG. 4. Spectral distribution of radiation from daylight 
fluorescent lamp (6500 k) . 

large floor-mounted units impractical and makes desirable the use of 
lighting equipment suspended from overhead. Hence, the more 
recent practice of using large numbers of reflector-type lamps. As 
mentioned in the introductory remarks, unfiltered tungsten produces 
considerable discomfort to the personnel at the illumination levels 
required for most present-day pickup devices, and large* numbers 
of reflector-type lamps are not well adapted to simple heat filtering 
methods. The lens-type motion picture studio spots employing the 
1-, 2-, and 5-kw lamps are much better adapted to the use of heat 
filters, particularly water cells, from 2 standpoints: (1) The fewer 
lighting units required, hence fewer filters, and (2) the 5-kw spot 
as used for motion picture studio lighting, for example, employs a 
14-in. diameter lens, making possible a relatively small water filter 



June, 1946 



APPRAISAL OF ILLUMINANTS 



437 



unit. It is therefore suggested that lighting equipment similar to 
that employed for motion picture studio lighting, using 2- and 5-kw 
lamps, and provided with proper suspension devices to permit them 
to be mounted overhead, be considered. Attention is directed to 
the dash-line curve of Fig. 3, which illustrates the theoretical re- 
sponse of the iconoscope to tungsten light after passing through a one- 
inch layer of water. The great reduction of infrared energy, partic- 
ularly beyond 9000 A is to be noted. 



8-40 WATT DAYLIGHT FLUORESCENT MAZDA LAMPS 




2345678 
FEET FROM CENTER LINE 

FIG. 5. 

However, the weak response from yellow, green, blue, and violet 
(6000-4000 A) as previously mentioned is quite apparent, hence 
the suggestion that the tungsten light be combined with that from 
mercury sources for better rendition of faces and set properties in the 
television image. 

Fluorescent Lamps. From the standpoint, of its spectral energy 
distribution, the daylight-type fluorescent lamp should be an excellent 
source for the illumination of television sets. Reference to Fig. 4 
shows a well-balanced energy output throughout the visible spectrum 
resulting in a satisfactory reproduction of all colors, including flesh 
tones, and a minimum of radiant heat. (Refer also to Table 1 in the 
Appendix.) 



438 



R. E. FARNHAM 



Vol 46, No. 6 



The outstanding difficulty with fluorescent lamps is their relatively 
low light output per unit area so that it is difficult to obtain the requi- 
site light levels, even if the studio were literally papered with lamps. 

This is borne out by reference to Fig. 5 which shows the illumina- 
tion at different distances in front of a bank of eight 40-w daylight 
lamps. Even if the bank is extended sidewise as well as vertically, 
the illumination at 6 ft, for example, in front of the lamps cannot be 
made to exceed approximately 350 ft-c. Substitution of larger lamps 




FIG. 6. A television studio (WRGB) employing water-cooled mercury 
lamps, for both overhead and floor lighting equipment. Each unit contains 
3 lamps operating on a 3-phase circuit. 


(100 w, 60 in.) will not appreciably alter this value inasmuch as the 

larger lamps operate at a light output per unit area of the same order 
of magnitude as the 40-w size. 

While future developments in fluorescent sources may make avail- 
able lamps of somewhat higher brightness, there is nothing at present 
that would indicate the attainment of 1200 ft-c at 6 ft. The general 
use of fluorescent lamps for television studios will have to await the 
adoption of much more sensitive pickup devices. 

REFERENCES 

1 BREEDING, H. A.: "Mercury Lighting for Television Studios," Proc. I.R.E., 
31,3 (Mar., 1943), p. 106. 



June, 1946 APPRAISAL OF ILLUMINANTS 

Appendix 

TABLE 1 

Energy Distribution of Possible Television Sources 
(Data given in per cent of input watts) 



439 



Source 



AH-6 Mercury 
A H-9 Mercury 
Tungsten 3350 K 
Daylight Fluor. (40 w) 



3165- 
3800 A 
Near UV. 


3800- 
5000 A 
Violet 
Blue 


5000- 
7600 A 
Green 
Yellow 
Red 


7600- 
14,000 A 
Near 
Infrared 


14,000- 
26,000 A 
Infrared 
(Heat) 


5.3 


12.1 


11.5 


6.1 





3.0 


4.6 


7.6 


6.1 


6.7 


. . . 


2.5 


14.9 


22* (40.0) 


0* (31.0) 


0.32 


6.7 


10.6 


0.47 


. . . 



* Based on use of water filters. Values in parentheses are without the filter. 



The removal of heat from light can be accomplished in 2 ways: (1) the use of 
glass heat-absorbing filters, such as Aklo glass, and (2) water cells. 

A typical Aklo filter will absorb 80 per cent of the radiant heat and 25 per cent 
of the light, thus giving definite benefit but requiring ( an appreciable increase in 
lamp wattage. The energy absorbed by the filter is radiated into the room making 
it necessary that the air conditioning system handle the entire lamp wattage in 
addition to its usual load. 

In the case of the AH-6 lamp, more than 50 per cent of the lamp's wattage is 
conducted out of the studio by its water-cooling arrangement. A water cell placed 
in front of the lens of an incandescent spot will absorb approximately 75 per cent 
of the infrared energy in the beam. This may amount to 15 per cent of the total 
lamp wattage and is conducted out of the studio, thereby lessening the load on 
the air-cooling equipment. 



(Q 

The grid of fluorescent lamps used to obtain the data of Fig. 5 employed 5-in. 
wide Alzak finish aluminum reflectors. Their contour was such as to give a slightly 
directional pattern to the light distribution. 

The bank of lamps was 40 in. wide and 48 in. high. The starting point of the 
light measurements was at the center of the bank vertically as well as horizontally, 
and the meter moved in a direction parallel to the plane of the lamps at the dis- 
tance indicated. 

Assume 9 similar banks arranged in 3 rows of 3 each. With the meter 6 ft in 
front of the center of the middle bank, the curves of Fig. 5 will show the following 
foot-candles to be received from each of the 9 banks of lamps : 



440 R. E. FARNHAM 

Center bank 54 ft-c 

Left and right banks with centers J45 ft-c 

40 in. to right and left \45 ft-c 



Top and bottom banks with centers 

48 in. above and below 
Four corner banks with centers 

62 1 / 2 in. away 



Estimated from lamps outside the above group 



38 ft-c 
38 ft-c 
26 ft-c 
26 ft-c 
26 ft-c 
26 ft-c 
26 ft-c 



350 ft-c 



CARBON ARCS FOR MOTION PICTURE AND TELEVISION 
STUDIO LIGHTING* 

F. T. BOWDITCH,** M. R. NULL,** R. J. ZAVESKYf 

Summary. Photometric and spectral energy distribution data are given for 
typical carbon arc light sources used extensively in motion picture studio lighting. 
The balanced color quality and low infrared content which make these sources valuable 
in motion picture photography are reviewed from the standpoint of the requirements 
of television studio lighting. The conclusion is reached that as television expands to 
justify adequate studio facilities, the carbon arc should find a place similar to that 
which it now holds in the motion picture studio. 

The motion picture industry utilizes carbon arcs not only as the 
light source for the projection of film in theaters but also for the 
illumination of sets during the filming of the picture. The types and 
applications of carbon arc units for studio lighting have been de- 
scribed previously before this Society. 1 The various spotlights and 
floodlights have been developed 2 ' 3 ' 4 as the industry grew, to meet 
special requirements for color, steadiness, light intensity, and freedom 
from noise. When sound was introduced in the industry there were 
noise problems which were solved by designing carbons, lamps, and 
associated equipment to operate quietly. Also, the advent of color 
in motion pictures made it necessary to use increased light intensities 
and to have light of the proper spectral quality. These needs were 
met by improved carbon arc units. 

Now that television is assuming greater commercial importance, 
problems associated with the lighting of television sets will demand 
increased attention.' As these sets become more elaborate and the 
photographic values become subject to more critical review, it seems 
probable that carbon arc studio lighting will find advantageous 
application here for much the same reasons that now give it an im- 
portant place in the motion picture studio. The purpose of this paper 
is to describe the operating and radiant energy characteristics of 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Research Laboratories, National Carbon Co., Inc., Cleveland, Ohio. 
f Development Laboratories, National Carbon Co., Inc., Fostoria, Ohio. 

441 



442 



BOWDITCH, NULL, AND ZAVESKY 



Vol 46, No. 6 



typical carbon arc flood and spot lamps which appear to offer worth- 
while advantages in television studio usage, as they do now in mo- 
tion picture studios. 

Motion Picture Studio Floodlight. The M-R type 40 Duarc 
lamp 3 - 4 illustrated in Fig. 1 is an example of a carbon arc flood lamp 
in common use in motion picture studios. Lamps of this type op- 
erate quietly and automatically. They are equipped to burn 2 arcs 
simultaneously, each with an 8-mm X 12-in. M. P. Studio positive 





FIG. 1. M-R type 40 Duarc lamp. 



and a 7-mm X 9-in. M. P. Studio negative carbon. The 2 arcs are 
in series and are operated at 40 amp d-c with approximately 37 v 
across each arc. The necessary ballast is incorporated within the 
housing, so that the unit need only be connected to a standard 110- 
v, d-c power line, to be ready for completely automatic operation. 
About 2 hr burning time is afforded by the 2 pairs of carbons. 

Fixed reflectors and a diffusing glass comprise the optical system 
used with the carbon arcs in this lamp, giving a spread beam which is 
used a great deal for general illumination in motion picture studios. 



June, 1946 



CARBON ARCS 



443 



The illumination characteristics of this flood lamp are depicted 5 in 
Fig. 2. For purposes of illustration the light data are reported for 
distances of 7 ft and 14 ft from the lamp. A peak value of 700 ft-c 
is obtained at 7 ft and a peak intensity of 165 ft-c at 14 ft. These 
peak intensities are obtained at the center of the beam; values for 
the other portions of the beam are as shown in Fig. 2. 

Motion Picture Studio Spotlight. A typical example of a carbon 
arc spot lamp in common use in motion picture studios is the M-R 



ouu 
700 

K 6 

_i 

3500 

K 

i? 400 

z 

|300 

H 

-200 
100 























I 


7 


-\ 


\ 










| 


A 

FEET r 


T 

*OM LAI 


\ 










/ 






\ 








I 










\ 






/ 


. 








\ 




-/ 


/ 

._ - 


. 

14 


_ 

A 

FEET FP 




T 
OM LAK 


- 
IP 


\ 
. 


V 








1 









8 



8 



4 2 0. 2 4 

RADIUS OF SPOT IN FEET 

FIG. 2. Horizontal distribution of light intensity for 
M-R type 40 Duarc lamp. 



type 170 lamp illustrated in Fig. 3. In this lamp 16-mm M. P. 
Studio positive and 1 /2-in. copper-coated M. P. Studio negative car- 
bons are burned at approximately 150 amp d-c. Units such as these 
are semiautomatic in operation and provide a burning time of the 
order of iy 2 hr with one pair of carbons. 

The studio spot lamps are equipped to give beams of different 
divergence by adjusting the position of a lens with respect to the 
light source. A 20-in. diameter Fresnel lens is used for the type 170 
lamp. The sketch in Fig. 4 illustrates the simple action of this system. 
The full lines in this sketch represent the lens position, the angle of 
light pickup and the beam divergence for maximum spread. By 



444 



BOWDITCH, NULL, AND ZAVESKY 



Vol 46, No. 



moving the lens farther away from the positive carbon crater a less 
divergent beam is obtained; the position of minimum divergence is 
shown in the sketch by the broken lines. 

The distribution of light intensity in the beam of the M-R type 170 

lamp is shown 6 in Figs. 5, 6, and 
7. Fig. 5 shows the light dis- 
tribution at 25 ft from .the 
lamp for lens adjustments pro- 
ducing 24-, 32-, and 48-deg 
beams. Figs. 6 and 7 show 
data at 50 and 75 ft, respec- 
tively. It is evident from these 
figures that the wide range of 
intensities and distributions ob- 
tainable provides a versatile 
source 'of illumination for sets 
of various sizes and for special 
lighting effects. 

Spectral Sensitivity Con- 
siderations. One of the most 
important characteristics of a 
source for studio lighting is 
the spectral distribution of the 
radiant energy in the beam. 
Motion picture film and tele- 
vision pickup tubes respond 
only to radiation of certain 
wavelengths. Radiation of 
other wavelengths falling on 
the set serves no useful pur- 
pose and is in fact a definite 
nuisance, since it is absorbed 
as heat by all objects which 

it strikes. Further, the relative spectral intensity of the source 
within the wavelength region to which the receiving device is sensi- 
tive is of importance in determining image densities with black- 
and-white recording, or a proper balance of colors if color reproduc- 
tion is desired. For these reasons, the spectral energy distribution 
curves of Fig. 8 are of particular interest. 

The color quality of the radiation of the type 40 Duarc has been 




FIG. 3. M-R type 170 lamp. 



June, 1946 



CARBON ARCS 



445 



found to be quite satisfactory, either alone or mixed with sunlight, in 
producing a proper color balance with Technicolor photography. 



20 INCH DIAMETE* 
FRESNEL LENS 







FULL FLOOD POSITIC 



MINIMUM SPOT POSITION 



FIG. 4. Optica^ system of M-R type 170 lamp. 

The direct radiation of the type 170 lamp, while very satisfactory 
for black and white, has been found relatively too intense in the 
blue and near ultraviolet when used with color, so that it is custo- 



1800 




12 8 4 4 8 

RADIUS OF SPOT IN FEET 

FIG. 5. Horizontal distribution of light intensity 
for M-R type 170 lamp (intensity at 25 ft). 

marily modified by the use of a yellow filter, placed over the front of 
the lamp, to produce a color quality almost identical with that of the 
type 40 Duarc. 



446 



BOWDITCH, NULL, AND ZAVESKY 



Vol 46, No. 6 



The infrared radiation from these sources has been measured by 
means of a series of filters 7 isolating particular wavelength regions of 

interest. These data are given in Table 1. 



TABLE 1 

Distribution of Radiant Energy at the Center of the Beam 

Percentage of Total in Each Wave Band 



Lamp 

Flood M-R-40 
Spot M-R-170 



For purposes of seeing, only the energy of wavelengths between 
0.40 and 0.70 ju is of any significant value. All the rest, approximately 
65 per cent of the total in this case, is endured for the sake of the 
useful portion. A similar situation holds with all sources, and the 



0.34- 
0.40 / 


0.40- 
0.70 fi 


0.70- 
1.125/t 


1.125- 
4.20^ 


4.20- 
12.0 M 


4 


34 


21 


34 


8 


4 


36 


28 


31 


1 



24 BEAM 
DIVERGE MCE;? 




8 4 4 8 12 

RADIUS OF SPOT IN FEET 

FIG. 6. Horizontal distribution of light intensity 
for M-R type 170 lamp (intensity at 50 ft). 

ratio indicated for these arcs is typical of a comparatively cool 
source as commercial units go. This will become apparent from the 
discussion to follow. 

In motion picture photography and in the television studio, the 
primary purpose of the light source is not to provide good vision, but 
to stimulate photochemical reactions on film in the first case and the 






June, 1946 



CARBON ARCS 



447 




12 8 4 4 8 12 
RADIUS OF SPOT IN FEET 

FIG. 7. Horizontal distribution of light intensity 
for M-R type 170 lamp (intensity at 75 ft). 




.5 .6 

WAVELENGTH IN MICRONS 



FIG. 8. Spectral energy distribution of studio light sources. 



448 



BOWDITCH, NULL, AND ZAVESKY 



Vol 46, No. 6 



photoelectric response of the mosaic of the pickup tube in the other. 
The spectral response of film and of the mosaic are thus the proper 
measures of light effectiveness, rather than the lumen or foot-candle. 
The photoeffectiveness of carbon arc sources with motion picture film 
has previously been discussed before this Society. 8 It is our purpose 
here to examine in similar fashion the application of carbon arc 
sources to television studio lighting, in order to obtain some indica- 
tion of their possible utility in this field. 




WAVE-LEN6T-H- iJ 

FIG. 9. Approximate spectral sensitivity of 
the Iconoscope: A Combined response of the 
Iconoscope with a light source consisting of a 
100-w projection lamp and a Wratten No. 78- A A 
filter; B Relative spectral sensitivity of the 
Iconoscope surface alone. 



As a prerequisite to such a consideration, some measure of the 
spectral sensitivity of the television pickup device is required. This 
information is shown by the curves of Fig. 9, which give an approxi- 
mate specification of the spectral sensitivity of the Iconoscope mosaic 
widely used in present-day television pickup cameras and occupying 
a position exactly analagous to that of the photographic film in the 
more conventional camera. 

Curve A of Fig. 9 reproduces the relative spectral response char- 
acteristic as published by the manufacturer "for an average Icono- 
scope when used with a light source consisting of a 100-w projection 
lamp operated at rated voltage, with a Wratten filter No. 18AA to 
approximate a source at 5500 K within the visible range." 9 Curve B 
of Fig. 9 has been calculated from the transmission characteristics 



June, 1946 



CARBON ARCS 



449 



of the filter and the spectral emissivity of the source to give the 
relative spectral sensitivity of the Iconoscope mosaic itself. This 
latter curve has then served as the basis for the calculations now to be 
described. 




FIG. 10. Comparative spectral intensities of light sources producing the same 
Iconoscope response, Pt. 1. 



RELATIVE SPECTRAL INTENSITV 
N> v>i -C O 










































~^ 


^ 


INCANDES 


:ENT LAMP 




















~~ _ 


^^_ 


~ 


-. 


-~_ 


a 








CARBON 4 


\RC 








U 2 !f tf 1-6 '7 1-9 20- 2 
WAVE -LENGTH -u 

FIG. 11. Comparative spectral intensities of light sources producing the same 



Iconoscope response, Pt. 2. 

Given the spectral sensitivity of the Iconoscope and the spectral 
intensity of a light source directed upon it, the product of the two, 
wavelength by wavelength, gives a measure of Iconoscope response 



450 BOWDITCH, NULL, AND ZAVESKY Vol 46, No. 6 

to direct radiation of this quality. Since reflected rather than direct 
light from the source reaches the Iconoscope in practical use, it is 
perhaps more logical to view such a product as the Iconoscope re- 
sponse to radiation reflected from a pure white object, (i.e., one hav- 
ing the same reflectivity at all wavelengths) when illuminated by the 
source under evaluation. In this present study, such calculations 
have been made for 3 sources commonly considered for television 
studio lighting: (a) the incandescent tungsten reflector spot lamp, 
perhaps the one most widely used in present-day experimental tele- 
vision studios, .(b) the type AH-6 water-cooled capillary mercury arc 10 
and (c) the carbon arc motion picture studio flood lamp previously 
described in this paper. Spectral intensity data for the first 2 sources 
were obtained through the courtesy of W. E. Forsythe and B. T. 
Barnes of the Lamp Development Laboratory, General Electric Com- 
pany, Nela Park, Cleveland, Ohio. 

Iconoscope response curves have been calculated for each of these 
3 sources, the ordinate at each wavelength being the product of the 
corresponding radiant energy and Iconoscope response. The areas 
under the 3 curves, proportional to the total Iconoscope response in 
each case, were then determined, and a suitable factor chosen for 
adjusting the intensity of each source, so that all 3 would produce the 
identical Iconoscope response. Figs. 10 and 11 show a comparison 
between the spectral intensities of the 3 sources after these factors 
have been applied to the original spectral energy distribution curves. 

These curves are a measure in each case of the spectral intensity 
of the radiation falling on a white object to produce a given Iconoscope 
response. Along with the useful radiation within the sensitivity 
limits indicated by Fig. 9, the curves of Figs. 10 and 11, particularly 
the latter, show the accompanying infrared energy, which is altogether 
useless. Since the area under the entire spectral energy distribution 
curve is a measure of the total radiant energy in each case, the reason 
for the considerable heat under high-intensity incandescent illumina- 
tion is quite apparent. If this energy be made the basis of compari- 
son at 100 per cent, then that of the carbon arc is 17.6 per cent and 
that of the mercury arc only 7.7 per cent, all giving the same total 
Iconoscope response. 

The energy advantage indicated for the mercury arc largely results 
from the fact that it is water-cooled, about 60 per cent of the total 
radiant energy 10 being carried off by the water. Similarly, a water- 
cell filter could be mounted in the beam from the carbon arc to remove 



June, 1946 



CARBON ARCS 



451 



approximately 50 per cent of the total radiant energy and so bring this 
source to approximate equivalence with the mercury so far as maxi- 
mum Iconoscope response per unit of radiant energy is concerned. 
With the incandescent spot lamp, the water cell would remove approx- 
imately 60 per cent of the total energy and so still leave this source 
about 5 times hotter than the carbon and mercury arcs. 

Color Response. The discussion so far has been concerned alto- 
gether with the achievement of maximum Iconoscope response per 
unit of radiant energy, without regard to color rendition. How- 




WAVE-LENGTH-y 



FIG. 12. Color response of Iconoscope, light source combinations 
adjusted to the same total response. 

ever, in practice, this cannot be ignored, as reference to the curves 
of Fig. 12 will show. These curves portray the color response of the 
Iconoscope with each of the 3 light sources under discussion, the 
response at each wavelength ordinate being the product of the 
Iconoscope sensitivity and the spectral intensity of the source at that 
wavelength. The curve for the incandescent spot lamp shows a 
maximum response to green, orange, yellow, and red colors of wave- 
lengths from 0.5 to 0.65 M and a significant response to reflections in 
the near infrared, of wavelengths beyond 0.7 ju. In contrast, violet 
and blue are less effectively reproduced than are the other colors. 
Since, as with black-and-white motion picture photography, these 
responses can only be effective in determining contrasts in a mono- 



452 



BOWDITCH, NULL, AND ZAVESKY 



Vol 46, No. 6 



chromatic image, the blues will appear unnaturally dark, the oranges, 
yellows, and reds too bright, and objects which are good infrared 
reflectors will assume a brightness to which the eye would not have 
responded at all in the original scene. Similarly, with the mercury arc, 
unnatural brightnesses of objects which reflect ultraviolet and violet 
are produced, with low responses in the blue-green, and particularly 
in the yellow and red. Responses with the carbon arc are somewhat 
more uniform, although showing a significant response to the near 
ultraviolet and a lower red as compared with th'e green and blue re- 
sponses. 




WAVE-LENGTH- 



FIG. 13. Iconoscope response with incan- 
descent tungsten lighting with and without 
color-correcting Wratten 78- A A filter. 

Distortions of this character require correction in practical usage 
in order to achieve natural density gradations in monochrome, or 
more importantly, if a color process is to be satisfied. With present- 
day monochromatic television, a type 78- A A Wratten filter is 
sometimes employed with incandescent lighting to give more natural 
contrasts. The use of such a filter greatly reduces the Iconoscope re- 
sponse, as the curves of Fig. 13 indicate. 

In this figure, the upper curve showing the color response for the 
incandescent source without filter is identical with that shown on 
Fig. 12, expanded vertically to illustrate better the loss encountered 
when the filter is employed. The lower curve of Fig. 13 shows the 
Iconoscope response with the No. 78- A A Wratten filter and the same 
source. The total Iconoscope response has fallen to one-sixth its 



June, 1946 CARBON ARCS 453 

former value, requiring that 6 times the intensity of incandescent 
lighting be applied to recover the loss in sensitivity introduced by 
the use of the filter. 

Thus, the necessary use of color-correcting filters can have a very 
profound effect on the radiant energy requirements for a television set, 
and calculations such as have been reported here can only be re- 
garded as indicative of experimental possibilities. However, if the 
better color response with carbon arc lighting suggested by Fig. 12 
indicates a correspondingly lower loss in correcting filters, then, in the 
final analysis, the relative efficiency and comfort of carbon arc light- 
ing should be significantly greater than that previously indicated 
in this paper. And, as the television industry grows to demand more 
elaborate stage settings and to achieve a higher quality of photo- 
graphic reproduction, the carbon arc should come to occupy an im- 
portant position in television just as it now holds in the motion picture 

studio. 

* 

REFERENCES 

1 LINDERMAN, R. G., HANDLEY, C. W., AND RoDGERS, A.: "Illumination in 
Motion Picture Production," /. Soc. Mot. Pict. Eng., XL, 6 (June, 1943), p. 333. 

2 RICHARDSON, E. C.: "Recent Developments in High-Intensity Arc Spot 
Lamps for Motion Picture Production," /. Soc. Mot. Pict. Eng., XXVIII, 2 (Feb., 
1937), p. 206. 

3 MOLE, P.: "The Evolution of Arc Broadside Lighting Equipment," /. Soc. 
Mot. Pict. Eng., XXXII, 4 (Apr., 1939), p. 398. 

4 JOY, D. B., LOZIER, W. W., AND ZAVESKY, R. J.: "Recent Improvements in 
Carbons for Motion Picture Studio Arc Lighting," /. Soc. Mot. Pict. Eng., XXXIII, 
4 (Oct., 1939), p. 373. 

6 Private communication from Mole-Richardson Co. 

6 Data calculated from Reference 2 above. 

7 COBLENTZ, W. W., DORCAS, M. J., AND HUGHES, C. W. : "Radiometric Meas- 
urements on the Carbon Arc and Other Light Sources Used in Phototherapy," 
Sclent. Papers oj the Bur. of Stand., 21,539 (Nov., 1926), p. 535. 

8 BOWDITCH, F. T., AND DOWNES, A. C. : "The Photographic Effectiveness of 
Carbon Arc Studio Light Sources," /. Soc. Mot. Pict. Eng., XXV, 5 (Nov., 1935), 
p. 375. 

9 RCA VICTOR DIVISION, RADIO CORPORATION OF AMERICA: "RCA 1846 Icono- 
scope," Descriptive Bulletin 1846-3-46. 

1(T BREEDING, H. A.: "Mercury Lighting for Television Studios," Proc. I.P.E., 
31,3 (Mar., 1943), p. 106. 



AN IMPROVED FILM-DRIVE FILTER MECHANISM* 
C. C. DAVIS** 

Summary. The basic problem of providing constant speed past the scanning 
point, in film machines is briefly reviewed. Two associated types of relatively simple 
film-drive filters are described. Both are based on the use of controlled compliance 
which promotes greater freedom from film and mechanical conditions. A new film 
sprocket is described which reduces the number of sprockets required per machine. 

It has long been recognized that speed variations seriously degrade 
the finest sound recordings. The harshness caused by high-frequency 
flutter and the "wows" caused by low-frequency flutter are well known. 
Less obvious perhaps is the fact that the selective nature of flutter 
with respect to the ear and to the program material may result in 
degradation without the appearance of flutter of appreciable mag- 
nitude. The public is fortunate that the motion picture industry has 
shown a continuing tendency toward more strict requirements for the 
performance of film propulsion equipment. Much has been written 
on the subject of flutter, frequency'modulation, or constancy of speed, 1 - 2 
and the various mechanical filter devices built into recorders and re- 
producers have been analyzed in detail. 2 " 5 These analyses are gen- 
erally made by converting the mechanical circuit into the equivalent 
electrical circuit since the theory of electrical wave filters is more 
familiar. Such analogies constitute a useful tool in the development 
and study of mechanical filters and will be used here to illustrate their 
functions. Since the solutions of such circuits appear in reference 
books they will be omitted. 

Review of Smooth Drum Filters. It is believed a description of 
the new equipment as well as a basis of comparison will be assisted 
by a brief consideration of the general type of film filter most 
commonly used at present. In earlier machines it was natural 
to copy the method of picture projection and record or reproduce on, 
or adjacent to, a film sprocket. Although a filter was included it 

* Presented Oct. 17, 1945, at the Technical Conference in New York. 
** Electrical Research Products Division, Western Electric Co., Hollywood. 
454 



FILM-DRIVE FILTER MECHANISM 455 

was ineffective against the large amount of flutter generated by 
sprocket and film pitch mismatch, so it was soon recognized this 
method had severe limitations. While it is necessary to synchronize 
a motion picture frame precisely with the adjoining sprocket holes, 
slightly more latitude is permissible in the case of the sound track. 
Thus the way was opened to ignore the sprocket holes at the trans- 
lation point and build a somewhat independent filter device as a 
sort of machine within a machine devoted to maintaining constant 
film velocity without regard to the instantaneous velocity of the 
remainder of the machine. 

In reviewing the principles used in film machines such as the com- 
prehensive reviews of Albersheim and MacKenzie 2 and Kellogg, 4 it is 
apparent more recent machines depend for their flutter attenuation 
upon the ability of a flywheel to revolve at constant velocity. This 
principle is utilized by rotating tlie flywheel by means of the film 
itself running as a belt over a smooth drum rigidly attached to the 
flywheel and with sufficient freedom or compliance introduced in the 
incoming and outgoing film paths to permit the flywheel to assume 
constant velocity. Nonuniformities of rotational motion in the 
driving mechanism are absorbed in the compliance of the film path, 
preventing them from altering the constant velocity of the massive 
flywheel. Damping must be included in such a combination to pre- 
vent continued oscillation or "hunting" of the flywheel which will 
inevitably result from the use of a free-running wheel compliantly 
driven. 

This smooth drum and flywheel filter arrangement, therefore, con- 
tains 3 basic elements: a flywheel and drum combination, an elastic 
or compliant loop of film, and a damping device. These are analogous, 
respectively, to an inductance, a condenser, and a resistor and are 
arranged in the form of an electrical low-pass filter passing low fre- 
quencies or essentially d-c components only. Theoretically these 3 
basic elements are sufficient to meet the most exacting flutter require- 
ments, since all that is necessary to obtain unlimited filtering is to 
establish a sufficiently low cutoff frequency which will be referred to 
as the frequency at which attenuation starts. Actually a filter com- 
bination composed of a flywheel of small enough proportions to satisfy 
starting requirements, combined with a comparably large enough com- 
pliance to filter the lowest flutter frequencies occurring in the usual 
driving source, requires careful design consideration. Difficulty lies 
not in the fact that there are only 3 elements but rather in improper 



456 . C. C. DAVIS Vol 46, No. 6 

physical values and particularly in mechanical imperfections, since 
flutter may be generated by rotational discrepancies such as occur in 
ball bearings, within the filter parts themselves. Furthermore, if the 
film does not travel precisely with the drum, momentarily borrowing 
the motion of the flywheel, the relative motion between the two not 
only results in flutter but may also permit the sound track to weave. 

The compliance element or elastance introduced between the 
driving sprocket and drum has in general been provided by the 
nature of the film itself in the form of ^-shaped loops. 2 ' 3 Thus if the 
loading tension of the film loop is light, the S-shaped bend caused by 
the film reaction against wrapping around the drum and sprocket in 
reverse directions, offers a relatively large value of compliance. 
This is the type of compliance used in many machines because of its 
simplicity. However, it offers no facilities for the introduction of 
damping, the alternative being to damp the rotating flywheel directly; 
but direct flywheel damping of sufficient magnitude to control oscilla- 
tions results in more constant d-c drag than the film will tolerate. 
Therefore, direct flywheel damping is applied from a member rotating 
at approximately flywheel velocity to provide incremental damping 
without steady state drag. This rotating member may consist of a 
gear-driven auxiliary drive, 4 or an auxiliary flywheel known as a 
kinetic scanner or rotary stabilizer. 2 - 3 A refinement of the latter 
type has been developed by Wente and Miiller. 5 Thus the usual 
practice has been to drive the drum simply by an S-shaped loop 
while damping was applied to the rotating flywheel by a relatively 
complicated mechanical arrangement. 

General Description of New Type of Filter. The objective in 
the development of the general design described in this paper was 
to combine the proved features of former equipment with new 
features which would prevent the occurrence of troubles, the exist- 
ence of which has been demonstated by previous experience. The 
inherent advantages of the smooth drum and flywheel are main- 
tained but it differs in other respects to a considerable extent. It 
consists of the combination of a relatively simple flywheel driven 
by deliberately tensioned film with damping applied to the tension- 
ing device. This combination fulfills 4 desirable requirements for 
good design: 

(a) Optimum bearing conditions of rotational parts, 
(6) Application of damping to a nonrotating element, 



June, 1946 



FILM-DRIVE FILTER MECHANISM 



457 



(c) A value of compliance largely independent of the varying quality or bent 
condition of the film, 2 

(d) Firm contact between film and drum. 

Problems in the utilization of considerable film tension without a 
corresponding reduction in loop compliance have been overcome by 
the design of units which produce low cutoff frequencies with flywheels 
of reasonable size. The drum and flywheel are propelled by film which 
is deflected from a straight path by a spring-tensioned idler arm and 
roller with which a damper is associated. This tensioned path of film 



SOFT ROLLER 



R2 

FLYWHEEL AND 
MAGNETIC DRAG 




SOUND DRUM 
AND FLYWHEEL 



SOUND SPROCKET 



INTERNAL FILTER CIRCUIT 

FIG. 1. Single-arm filter. 



together with the idler arm and roller constitutes a compliance as in 
the case of the 5-shaped loop and will be referred to as a film loop 
although the two appear considerably different. 

Since substantially equal amounts of tension in some form must 
exist at either side of the translation point, tension must be intro- 
duced to offset that of the lower film loop. Two methods have been 
developed to accomplish this for 2 different types of application. 
The first type utilizes the eddy current drag created by the use of 
permanent magnets mounted near the flywheel periphery. This will 
be referred to as the single-arm type. The second type utilizes an 
idler arm and roller on the incoming film loop similar to that in the 
outgoing film loop. This will be referred to as the double-arm type. 



458 



C. C. DAVIS 



Vol 46, No. 6 



Thus, in either type the flywheel and drum are rotated by a ten- 
sioned lower film loop which deflects an arm into normal operating 
position at normal film speed. Discrepancies in velocity introduced 
into the film by the driving sprocket tend to deflect the relatively light 
and compliant arm from normal position rather than introduce flutter 
into the relatively massive flywheel. Any tendency to oscillate, 
which would appear as an up and down motion of the arm accom- 
panied by an acceleration and deceleration of the flywheel, is damped 
by the dashpot associated with the arm. 



-10 



0.25 



RELATIVE FREQUENCY F/FR 
1.0 2.0 40 60 16.0 



32.0 



64.0 



Z 20 

JO 
30 



40 



50 



FIG. 2. Attenuation characteristic of single-arm filter. 



Single- Arm Type. The mechanical schematic of the single-arm 
type is shown in Fig. 1. This type, using magnetic drag, produces 
a completely slack or free upper film loop because of the pressure 
roller which holds the incoming film against the drum. Since a free 
upper loop isolates the drum from disturbances occurring in the 
incoming film, this type is particularly applicable to theater repro- 
ducers because of the isolation it provides from disturbances in the 
projector head. 

The analogy of the filter circuit shown in Fig. 1 consists of a shunt 
path containing the flywheel and drag and a second shunt path 
containing the arm and damper in parallel with the film compliance. 
The 2 shunt paths are fed in parallel by a sprocket or generator. 



June, 1946 



FILM-DRIVE FILTER MECHANISM 



459 



Minor elements which produce no appreciable effect on the attenua- 
tion characteristic are omitted. The a-c attenuation is equal to the 
ratio of the disturbing velocity to that reaching the flywheel, or the 
ratio of current through the generator to the current through the 
shunt path containing the flywheel. The sprocket is shown as an 
infinite impedance or constant current generator since it is not af- 



LIGHT BEAM 



COMMON SPRING' 
Cl 




UPPER SPROCKET 

S2 



OIL DAMPER 

R 



SOUND DRUM 8, FLYWHEEL 
Ml 

C5 



LOWER SPROCKET 

SI 



Ml 




r^ 


|M2 


M3 


S2 


* 


hC5 |R 


= 


=C6 


^ 


*C2 


T C3 


Z oo I 






/ T CI 



^- INTERNAL FILTER CIRCUIT 
FIG. 3. Double-arm filter. 



fected by occurrences in the filter. The magnetic drag R 2 is only of 
sufficient magnitude to produce the desired d-c film tension and af- 
fords no appreciable damping. The mass of the filter arm M 2 has but 
slight effect on the filter characteristic owing to the limiting effect of 
RI. The film compliance Cz results from the small bends at points of 
contact with the tensioned film loop and is much smaller than the 
compliance of the arm C\. 

Fig. 2 shows an attenuation curve for the circuit of Fig. 1. The 
attenuation of 2 : 1 or 6 db per octave which results from the fixed 



460 C. C. DAVIS Vol 46, No. 6 

value of RI combined with an increase of 6 db per octave in the reac- 
tance of MI, increases to 12 db per octave at higher frequencies when 
the reactance of C 2 becomes less than R^ This fact is not particularly 
important since experience shows the flywheel mass is sufficient in 
any event to take care of high-frequency flutter and that most 
troublesome flutter frequencies are those occurring in the region where 
filtering is least and where the ear is most sensitive to flutter. Typical 
curves show the latter to be in the neighborhood of 2 cycles. 2 There- 
fore, the cutoff frequency and the shape of the filter curve in the vicin- 
ity of this frequency are of primary importance. The general 
characteristic of the present filter is desirable in that the resonant rise 
is less than 2 db and a low cutoff frequency is provided. 

Double-Arm Type. In Fig. 3 is shown the double-arm or alter- 
native method of applying the damped compliance flywheel drive. 
It differs from the previous type in that the magnetic drag and pres- 
sure pad roller are replaced by an idler arm and plain roller similar 
to that used in the lower film loop. This makes possible a single 
unit containing the arms and damper which perform all the functions 
necessary to drive the drum and flywheel by straightforward belt 
action. The elimination of the magnetic drag and the pressure pad 
roller is desirable from a standpoint of simplification and of more 
convenient and readily controlled operating conditions. Since the 
operating parts consist of plain rollers and a simple flywheel as- 
sembly, desirable conditions for film propulsion without contact 
with the sound or picture emulsion, and without weaving or danger 
of runoff, are inherent in the design. 

The practical application of the double-tensioned loop drive de- 
pends upon a design which supplies the relatively large amount of 
film tension best suited for flywheel starting and driving while at the 
same time supplying the amount of loop compliance necessary for a 
low cutoff frequency. The loops are not tensioned independently 
but by a differential action, through a single linking or common spring. 
Tensioning is done by the arms whose reaction to flywheel motion is 
to move the spring supports in the same direction, thus avoiding a 
proportionate change in the length of the spring. The only reactance 
opposing flywheel motion is that resulting indirectly from changes in 
the working angles through which the spring tension is applied to the 
loops. This arrangement permits more actual compliance with 2 loops 
than would result from a single comparable loop, thus facilitating the 
combination of a small flywheel and low cutoff frequency. 



June, 1946 FlLM-DRIVE FlLTER MECHANISM 461 

The circuit analogy of Fig. 3 includes the additional arm and addi- 
tional sprocket since the latter is not isolated by a free upper loop as 
in the single-arm type. These dual elements are shown separately 
in order to include the condition wherein a disturbance is introduced 
by either individual sprocket. Most significant flutter disturbances 
originate in both sprockets simultaneously and in phase, in which 
case' the circuit may be combined into one similar to that previously 
described. 

The analogy of the common filter spring may require explanation. 
What might be referred to as its actual value of compliance is sfyown as 
Ci whose value is that seen by the drum through either arm with the 
other arm stationary and is therefore a function of the spring constant. 
The compliance values, 2 and C 3 , are seen by the scanner in the re- 
spective arms when the arms move together without causing appre- 
ciable change in spring length. Therefore, C 2 and C 3 are functions of 
the angles through which C\ works combined with the loading tension 
of Ci. It is interesting to note that if the arms and loops were in- 
finitely long, no operational changes in angles would occur and Cz 
and Cz would be infinitely large. This would indicate the internal 
filter circuit would be nonoscillatory and have an infinitely low natural 
period. Therefore the cutoff frequency and the natural oscillatory 
period of the filter circuit are functions of the geometry of the arms 
and loops and the loading tension of the spring and are not directly a 
function of the spring constant. This is of practical assistance in the 
design of a compact double-arm unit. 

A disturbance from a driving source, such as the motor or the main 
drive gearing, or from varying load conditions in any portion of the 
machine, is automatically transmitted to both sprockets simultane- 
ously by the intermediate gearing between sprockets. An individual 
sprocket may contain a disturbance originating in the intermediate 
gearing between sprockets. These 2 types of disturbances constitute 
2 types of drive side input; dual sprocket input and single sprocket 
input. 

When a dual sprocket input occurs both sprockets accelerate and 
decelerate simultaneously relative to constant sprocket velocity 
indicating in the analogy each passes the same current. Therefore the 
branch through Ci carries no current and has no effect on the circuit 
% if the small film compliances C 5 and C 6 are considered equal or ignored. 
A dual sprocket input then represents a current input equal to that 
generated by either sprocket alone superimposed upon the filter, with 



462 



C. C. DAVIS 



Vol 46, No. 6 



the same general attenuation characteristic as the single-arm type 
shown in Fig. 2. 

Fig. 4 shows the attenuation curve of the double-arm filter mecha- 
nism. Disturbances arising in individual sprockets produce individual 
attenuation curves because of the application of damping to one arm 
only. In general the attenuation curve resulting from disturbances 
introduced by Si associated with the damped arm is similar to that of 
dual sprocket input. Disturbances from 6*2 receive considerably more 
attenuation because the reactance of the arm is not limited by the 



-10 



0.25 



RELATIVE FREQUENCY F/FR 
1.0 2.0 4.0 6.0 I6.O 



32.0 64.0 




30 



'Si &S2 



5z ONLY 



\ 



\ 



FIG. 4. Attenuation characteristic of double-arm filter. 



resistance of the damper. This attenuation is especially large 3 or 
4 octaves above cutoff where the reactance approaches zero. 

New Sprockets. Since a film does not pass over a sprocket as a 
perfect belt, additional disturbances exist in the movement of the 
film in the filtered film path, the most serious of which has been 
called "cross-over." This condition results from alternate shifting 
of the free play between sprocket holes and sprocket teeth, usually 
forward and backward as a result of surging load conditions of film 
reels or take-ups. These varying inertia loads, whether or not com- 
bined with varying take-up friction, cause cross-over unless a free 
loop exists on one side of the sprocket. The usual procedure is to 
add an additional holdback sprocket with a free intervening loop of 



June, 1946 



FILM-DRIVE FILTER MECHANISM 



463 



film. A solution has been devised to avoid the inconvenience of the 
additional sprocket as well as to maintain the smooth characteris- 
tics of sprocket operation with tensioned film in either direction. 

This solution is a sprocket having as great a tooth width as will 
freely pass the 2 extremes of film pitch for a given set of operating 
conditions. Thus a 16-tooth sprocket with 74 mil teeth and with 90 
degrees wrap will propel film ranging in pitch ==0.4 per cent and the 
greatest free play will be 4 mils as compared to 20 or more mils with a 
standard sprocket. The wide. tooth combined with a large tooth 



0.25 0.5 



RELATIVE FREQUENCY F/F R 
IX) 2.0 4.0 6.0 16.0 32.0 64.0 



DAMPED COMPLIANCE SYSTEM 
WHERE 1 C = COMPLIANCE OF ARM 
C = COMPLIANCE OF FILM 
R= VALUES SHOWN 




5. Comparison of damped compliance and damped inertia 
systems. 



radius is convenient to thread and results in a sprocket operating 
equally well in a feed or pull-down position with standard positive 
film stock. 

Transients and Load-Side Disturbances. The transient response 
of the single-arm and double-arm filters is that of a series resonant 
circuit and may therefore be computed on the basis of similar elec- 
trical circuits. Actually, while increased damping decreases tran- 
sient response, optimum flutter performance results from an amount 
of damping determined by recurring disturbances. Disturbances 
caused by the transient nature of film splices are minimized by the 
relatively well-damped nature of the circuit and by the fact that 



464 C. C. DAVIS 

there is little change in compliance in the presence of large dis- 
turbances. 

Load-side disturbances or those originating within the filter circuit 
proper, such as ball-bearing disturbances, are introduced into the 
internal filter circuit and are attenuated in proportion to the imped- 
ance of this circuit. Therefore the attenuation is minimum at the 
natural resonant frequency of the flywheel and arm combination. 

A comparison between a typical kinetic scanner and a solid wheel of 
equal total proportions and driven by equal compliances is shown in 
Fig. 5. The kinetic scanner is referred to as the damped inertia system 
and the solid wheel as the damped compliance system. An inspection 
of the curves shows that the damped compliance system offers more 
attenuation in the low-frequency range where the ear is most critical 
to flutter. Three values of damping are plotted for the damped com- 
pliance system, one-half critical damping representing a typical value. 

The application of values in rotational mechanical units to the cir- 
cuits of Figs. 1 and 3 has facilitated development of the single- and 
double-arm filters. Substantial agreement was obtained between 
computed values of elements and those obtained by physical measure- 
ments. The resulting filter characteristics were in turn checked by 
graphic charts on a laboratory flutter measuring instrument 6 which 
also provided the means for measuring flutter performance. Applica- 
tion of the foregoing methods has made it possible to predict the de- 
sign performance of recorders, rerecorders, and reproducers with less 
dependence on the experimental approach than has generally been re- 
quired in the past. 

REFERENCES 

1 SHEA, T. E., MACNAIR, W. A., AND SUBRIZI, V.: "Flutter in Sound Records," 
/. Soc. Mot. PicL Eng., XXV, 5 (Nov., 1935), p. 403. 

2 ALBERSHEIM, W. J., AND MACKENZIE, D.: "Analysis of Sound-Film Drives," 
/. Soc. Mot. Pict. Eng., XXXVII, 5 (Nov., 1941), p. 452. 

3 COOK, E. D.: "The Technical Aspects of the High-Fidelity Reproducer," 
/. Soc. Mot. Pict. Eng., XXV, 4 (Oct., 1935), p. 289. 

4 KELLOGG, E. W.: "A Review of the Quest for Constant Speed," /. Soc. Mot. 
Pict. Eng., XXVIII, 4 (Apr., 1937), p. 337. 

5 WENTE, E. C., AND MULLER, A. H.: "Internally Damped Rollers," /. Soc. 
Mot. Pict. Eng., XXXVII, 4 (Oct., 1941), p. 331. 

6 SCOVILLE, R. R.: "A Laboratory Flutter-Measuring Instrument," /. Soc. 
Mot. Pict. Eng., XXIX, 2 (Aug., 1937), p. 209. 



A SIMPLIFIED ALL-PURPOSE FILM RECORDING MACHINE* 
G. R. CRANE AND H. A. MANLEY** 

Summary. This paper describes a completely new recording machine for 16- 
or 35-mmfilm having a minimum of weight and bulk with a maximum of simplicity 
and excellent basic performance. A new sealed light valve is used in a simplified 
modulator and optical system employing a prefocused lamp for variable-density re- 
cording. A microscope is available for observing the valve in the modulator, and 
photocell monitoring and other accessories may be added if desired. 

Recorder. The R A -1231 film recording machine described in 
this paper has been designed for the purpose of providing the in- 
dustry with a versatile machine for recording any of the standard 
original or release type of sound tracks on either 35-mm or 16-mm 
film. 

The recorder consists of a base on which is mounted the film com- 
partment, modulator housing, main drive motor and the necssary 
plug connectors, relay and motor disconnect switch. All other con- 
trols are external. The machine is shown in its standard form by Fig. 
1 which is a front view with the doors open showing the inside of the 
film and modulator compartments. Fig. 2 is a rear view of the re- 
corder with the cover removed showing the drive motor, chain drive, 
hand wheel, and plug connectors. The total weight of the recorder, 
less the film magazine, is approximately 76 Ib. 

A footage counter is provided on the film compartment door. 
Space has also been provided for additional accessories such as a 
photographic slater and film punch. A simple lever lock has been de- 
signed to retain the film magazine on the machine which reduces the 
time necessary for the operator to change magazines. 

The machine will accommodate either synchronous or ac-dc 
interlock motors. The drive from the motor to the main drive shaft 
of the film pulling mechanism is by silent chain. In this drive as- 
sembly, any of the normal motor drive speeds that are used in the 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Electrical Research Products Division, Western Electric Co., Hollywood. 

465 



466 G. R. CRANE AND H. A. MANLEY Vol 46, No. 6 

industry can be accommodated, by a simple change of chain sprockets. 
The film drive unit as seen in Fig. 3 is a complete assembly that is 
readily removable from the film compartment. For converting to a 
16-mm film recorder a similar unit completely assembled and tested 
may be installed in its place in the machine. The 2 film sprockets 
differ in tooth design from the standard film sprockets that are in use 
today. Their design is discussed elsewhere in the JOURNAL. 1 These 
2 film sprockets are driven from the main drive shaft through right- 




FIG. 1. The RA-1231 film recording machine, showing front doors open. 

angle helical gears, the pinion gear being steel and the large gear being 
nonmetallic to reduce mechanical noise. 

In this same drive unit is located an adjustable multiple disk clutch 
for the film magazine take-up drive. The take-up from this clutch to 
the film magazine is accomplished through a small F-belt. Using a 
F-belt working through the multiple disk clutch assures a relatively 
smooth take-up. For those who prefer the permanent arm pin drive- 
type magazine take-up in place of the F-belt drive the machine has 
been so designed that it may be added as an accessory. Mounted on 
the front of the film drive unit plate, between the 2 idler rollers and 
extending down and around to the right-hand film sprocket, is the 
antibuckle trigger mechanism. This trigger operates a motor "dis- 



June, 1946 ALL-PURPOSE FlLM RECORDING MACHINE 467 

connect" relay that locks itself in electrically, so that once the anti- 
buckle trigger is tripped the motor is disconnected and cannot become 
operative until the main motor disconnect switch is turned off, which 
allows the relay to drop back to normal operate position. A buzzer 
connected across the coil of the relay gives an audible signal in the 
event of a film buckle. A hand wheel is attached to the end of the 
main drive shaft and extends through a hole in the right-hand end of 
the main case. 

The recording drum unit, as seen in Fig. 3, is also a complete as- 
sembly that is readily removable from the main case of the recorder. 




FIG. 2. Rear view of film recorder, with 
cover removed. 

It consists of the recording drum, shaft, flywheel and support casting. 
The drum and flywheel are rigidly coupled to the shaft which is ball- 
bearing mounted in the support casting. 

The film drive filter mechanism, as seen in Fig. 3, is a compact 
unit assembly that is readily replaceable or may be converted to 
accommodate 16-mm film. This type of filter is described elsewhere 
in the JOURNAL. 1 The filter mechanism was designed for this ma- 
chine with the view toward keeping the film compartment clear of 
obstructions and easy to thread as shown in Fig. 1 . Only the 2 filter 
rollers extend into the film compartment. These rollers are flanged 
and guide the film onto the drum in the proper position for recording. 
No appreciable amount of weave has been encountered in using this 



468 G. R. CRANE AND H. A. MANLEY Vol 46, No. 6 

method of guiding. The dash pot, whose plunger is attached to the 
upper filter arm, is made of transparent plastic and an opening is 
provided in the mounting plate so that the fluid level is readily 
visible from the film compartment. The dash pot has been so de- 
signed that it will not lose fluid even if the machine is turned upside 
down. Film threading indicators are provided and are visible from 
the film compartment also. 

The recorder, as seen in Fig. 1, is provided with a small modulator; 
however, the section of the recorder which houses the modulator is 
removable and may be replaced by appropriate housings for other 
more complex modulators. 





FIG. 3. Filter unit, drive unit, and recording drum unit. 

Modulator. The modulator for this recorder has been designed 
to provide the simplest possible assembly containing the light valve, 
the lamp, and the optical system required to record either 16- or 
35-mm single variable-density track. Photocell monitoring and 
other accessory devices frequently found in modulators have been 
eliminated in the interests of simplicity, but there has been no 
compromise with the quality of the items and performance achieved 
with regard to either the light valve or the optical system. Photo- 
cell monitoring can be added as an accessory if desired. 

As shown by Fig. 4, the modulator consists of a chassis which is 
intended to be screwed onto a flat surface of the recorder and con- 
tains all of the necessary elements in one assembly. It is provided 
with a milled slot in the mounting surface which rides on a key at- 
tached to the film recorder case so that the recorded track position 



June, 1946 ALL-PURPOSE FlLM RECORDING MACHINE 469 

may be laterally adjusted without disturbing other adjustments. 
The range of lateral movement is sufficient to permit recording on 
either edge of 16-mm film and the standard position on 35-mm. 
Provision can also be made for recording on the opposite side of 35-mm 
film if required. In this application there is no cover over the modu- 
lator since the recorder case provides this facility, but if it is used for 
other applications a suitable cover is available. This modulator differs 
from its predecessors with respect to the 4-amp lamp, improved optics 
of the anamorphic type, the use of a sealed permanent magnet light 







FIG. 4. Modulator unit. 

valve, and facilities for mounting a special microscope to observe the 
ribbons without removing the valve from the modulator. 

The condenser lens system is designed to accommodate the rela- 
tively short and large diameter filament of the standard 9-v, 4-amp 
prefocused theater lamp. This lamp is chosen because of its wide 
use and availability in all parts of the world and also because of the 
fact that the short, heavy filament is rugged and not subject to de- 
formation with use. The prefocusing feature is also of value, since it 
reduces the lamp adjustments necessary to a single vertical screw 
adjustment which can be eliminated without appreciable loss of 
light or introduction of lamp noise. One of the condenser lens ele- 



470 G. R. CRANE AND H. A. MANLEY Vol 46, No. 6 

ments is mounted in the end of the light valve and therefore serves 
to seal up one end of the valve. A small subassembly containing this 
lens also contains the effective aperture of the valve which deter- 
mines sound track width, so that by changing this subassembly, the 
track width may be altered. This permits any valve to be converted 
from 35-mm, 100-mil track to 16-mm, 80-mil track by changing this 
item only from the outside of the valve. 

The objective employed is a high-quality achromatic lens and 
works in combination with a short focus cylindrical lens located near 
the film. The light valve ribbons are brought to a focus on the film, 
giving an effective slit height of approximately 0.25 mil for a 1.0-mil 
valve. The objective and the small auxiliary cylinder are mounted 
in such a manner that the azimuth of the cylinder is separately ad- 
justed and fixed, permitting the focus to be adjusted in the usual 
manner. These adjustments are controlled by 2 knurled rings which 
are located in the film compartment when the modulator is assembled 
on the recorder. The light valve azimuth is not adjustable, but it is 
accurately controlled in relation to the modulator and the recorder. 

Just ahead of the surface on which the light valve is mounted, a 
pair of grooves is provided which accepts an inspection microscope, 
described later in this paper. 

Light Valve. The light valve used in this modulator has been 
coded RA-1241-A for 35-mm, 100-mil track, and RA-1241-B for 16- 
mm, 80-mil track. These 2 valves are basically identical and differ 
only in the masking device which determines the track width. This 
valve, shown by Fig. 5, is the sealed permanent magnet type, approxi- 
mately l 5 /s in. sq and 15 /ie in. thick. It is 2-ribbon, bi-planar and is 
provided with silver contacts at the front edge of the rear surface. 
In the modulator it is held against a reference surface by 2 springs and 
is located by means of 2 dowel pins anchored in the modulator. 
The front surface of the valve is provided with ramps so that it may be 
pushed in easily and lift itself over the dowel pins while entering. 
The springs cause it to snap in place over the pins and it is readily 
removed by pressing the valve back against the springs, so as to clear 
the pins, and then pulling forward. As previously mentioned, the 
rear of the valve is sealed by a condenser lens and the front surface is 
sealed by a cover glass so that the interior is effectively sealed against 
dirt or magnetic particles. It is not intended that the valve should be 
opened by anyone except the manufacturer unless equipped with the 
proper type of magnetizing equipment, since the valve is magnetized 



June, 1946 ALL-PURPOSE FlLM RECORDING MACHINE 471 

after it is assembled and cannot be readily pulled apart without 
damage. 

This valve is normally tuned to 9000 cycles and in view of its small 
size, every effort has been made to produce the highest possible flux 
density in the air gap which results in good sensitivity and a relatively 
low resonance peak as compared to other valves of similar size. When 
tuned to 9000 cycles the average sensitivity is approximately +7.0 
dbm. The rise at resonance is approximately 12 db when the valve is 
used in the usual simplex circuit with a matched impedance trans- 
former. The circuits employed in associated equipment provide 
facilities for feeding the valve through a circuit containing a correc- 




FIG. 5. Light valve, seen from front and rear. 

tive, constant impedance equalizer having a frequency characteristic 
which is the inverse of the light valve. In addition, limiting facilities 
are provided which prevent serious overloading of the valve and ex- 
perience to date with this type of equipment has indicated the light 
valve to be quite stable and dependable over considerable periods of 
time. 

Microscope. The light valve microscope developed for this 
modulator is shown by Fig. 6. The purpose of the microscope is to 
provide a convenient tool for observing the valve ribbons without 
removing the valve from the modulator. This provides a facility 
for checking ribbon modulation, noise reduction, ribbon spacing, 
and azimuth. The microscope consists of a lightweight dural case 
attached to a steel tongue which is inserted in the slots provided in 
the modulator. Two small ball pressure points insure registration 



472 G. R. CRANE AND H. A. MANLEY Vol 46, No 6 

with one reference guide surface which guides its movement back 
and forth to check ribbon azimuth. This movement is aided by the 
use of a small lever located on the side of the microscope. When it 
is set in the position toward the operator, the microscope may be 
pushed in to a point where a stop is engaged and which corresponds 
to optical registration with one end of the light valve slit. Then 
by operating the lever to the vertical position where a detent is felt, 
the center of the light valve is in register and by moving the lever to 
its forward position the other extreme end of the valve may be 
viewed. For quick checks the lever may be left in the center posi- 
tion. 




FIG. 6. Light valve microscope. 

The optical path is fundamentally simple and consists of a 32-mm 
microscope objective to view the ribbons through a small prism which 
turns the beam 90 deg between the light valve and the objective. In 
order to maintain a reasonable working distance for the objective and 
yet obtain a 10 to one magnification at the eyepiece, a long rear focal 
distance is used and this is obtained by folding the optical system 
twice by means of totally reflecting prisms. The objective is adjust- 
able for focus and is controlled by a small focusing knob just below 
the eyepiece. The eyepiece is a 15-power hyperplane equipped with a 
reticle which contains 2 concentric circles for comparison with the 
image of the light valve ribbons. These 2 circles are of such size that 
their diameters correspond to a valve spacing of 0.8 and 1.2 mils, 
respectively. They are located off center in the reticle so that by a 



June, 1946 ALL-PURPOSE FlLM RECORDING MACHINE 



473 



rotation of the eyepiece the operator may bring the reticle pattern 
into register with the valve slit image. It has been found that an 
operator can quickly judge the valve spacing in terms of this pattern. 
The modulator cannot, of course, be operated for recording while the 
microscope is in place but it may be quickly inserted between takes 
if a check on valve performance is desired. 

Conclusion. The design of the complete recorder appears to be 
very satisfactory with regard to its simplicity of mechanical parts 
and freedom from critical adjustments. The unit assembly con- 



15 



TOTAL 



TER 



06 
.04 

.02 



10 
FLUTTER FREQUENCY BANDS 

FIG. 7. Flutter values. 



too 200 



struction of the film propulsion mechanisms has reduced servicing 
and maintenance considerably. A clear film compartment and the 
elimination of unnecessary controls facilitate the operation of the 
machine. 

Studies of flutter performance made to date indicate that it com- 
pares very favorably with the highest quality film recording machines 
in use today. As seen in Fig. 7, the total flutter for all frequency 
bands from 2 to 200 cps is approximately == 0.05 per cent. The amount 
of 96-cycle flutter caused by sprocket hole disturbances is essentially 
negligible. The low-frequency flutter has been reduced to values that 
are not audible to the ear in high-quality music reproduction. 



474 G. R. CRANE AND H. A. MANLEY 

It is believed that the over-all performance of the recorder com- 
bined with its simplicity and versatility will amply meet the long felt 
needs of the industry for this type of film recording machine. 

REFERENCE 

1 DAVIS, C. C.: "An Improved Film-Drive Filter Mechanism," J. Soc. Mot. 
Pict. Eng., 46, 6 (June, 1946), p. 454. 



THE USE OF DESICCANTS WITH UNDEVELOPED PHOTO- 
GRAPHIC FILM* 

C. J. KUNZ AND C. E. IVES** 

Summary. With the packaging materials now available it is possible for the 
manufacturer of photographic film to provide almost complete protection from un- 
favorable atmospheric humidities up to the time of use. Management of the film 
moisture content after the moisture impervious package is opened involves adoption 
of suitable techniques and in some cases the use of desiccants. 

Several materials of outstanding interest as desiccants have been studied as to 
moisture capacity and rate of absorption under suitable conditions. The rate of 
moisture exchange as influenced by disposition of the film, the presence of membranes, 
circulation of air, maintenance of reduced atmospheric pressure, etc., has been 
measured. 

Recommendations are made as to methods of minimizing moisture uptake, choice 
of desiccant, care of desiccant, utilization of material available in the field, and avoid- 
ance of the effects of over drying. 

Technology has made its greatest strides in regions of temperate 
climate and its products are, for the most part, fitted to use in such 
regions. Difficulties are frequently encountered, therefore, when 
they are used in other regions such as the humid tropical belt and so 
it is not surprising that a material designed to respond to forces so 
slight as a momentary exposure to light should be affected by heat 
and humidity. While photographic science has endowed these mate- 
rials with increased speed and more finely balanced properties without 
disproportionate sensitiveness to atmospheric conditions, the photo- 
graphic literature gives ample warning of the risk of speed loss and 
latent-image fading incurred when conditions are extreme. The 
present work was undertaken to obtain specific information as to 
measures which would be effective in the management of humidifica- 
tion and dehumidification when applied to the needs of the film user. 
Such measures may be required for the undeveloped film either before 
or after exposure. 

* Presented Oct. 16, 1945, at the Technical Conference in New York. 
** Eastman Kodak Company. Rochester. N. Y.; Communication No. 1069. 

475 



476 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

On occasion, these measures may also be applied in conditioning 
processed film; for instance, dehumidification may be used to re- 
tard fungus growth or to eliminate stickiness which would prevent 
motion picture film from traveling smoothly through the projector. 
The use of desiccants with processed motion picture film is not gener- 
ally advisable however, because of the danger of embrittling the film 
if desiccation is carried too far. Effective and safer procedures are 
known but consideration of them is outside the scope of the present 
paper. Humidification, on the other hand, may be required to over- 
come a tendency toward brittleness resulting from storage under ex- 
tremely dry conditions. While the same general principles as set 
forth here for the case of undeveloped film apply to developed film, 
the situation will differ quantitatively with the latter. 

Condition of Film. For the present purpose it will be sufficient 
to consider that raw film as supplied by the manufacturer has, 
in general, a moisture content which permits it to withstand, for 
reasonable periods before and after exposure, temperatures around 70 
F and middle-range humidities without serious loss of speed or de- 
terioration in image quality. It will be assumed here that the ab- 
sorption of additional moisture is a contributing factor in deteriora- 
tion, especially if the film is to be kept at high temperature for any 
considerable time previous to development. 

Climatic Conditions. Available information indicates that film 
may encounter temperatures from 80 F to +140 F, and relative 
humidities from a few per cent to practically 100 per cent. The 
high humidities are not ordinarily met at temperatures above 100 F. 
In this paper attention is centered on the control of film moisture 
content and temperature is considered only incidentally. 

Problem of Moisture Control. To the extent that facilities per- 
mit, recommendations as to temperature can be followed success- 
fully by using a thermometer and by exercising ordinary care and 
good judgment. In contrast, there is no such simple instrument by 
the use of which the moisture content of film, desiccants, etc., can 
readily be determined. Estimates of the amount of moisture taken 
up by film and desiccants, as well as of the influence of a number of 
factors on the rate of moisture transfer, may have to be made in the 
field with reasonable accuracy but without the aid of any instruments. 
Because of the lack of information, -opportunities for shielding the 
film from unfavorable treatment may be overlooked and measures 
adopted for desiccation may be unsuitable. Indeed, it now appears 



June, 1946 



USE OF BESICCANT s 



477 



doubtful that -much, if anything, has been accomplished when some 
of the procedures suggested in the literature were followed. 

PSYCHROMETRIC PRINCIPLES 

Photographic film, like paper, cotton, silk, leather, and many other 
organic substances, may contain several per cent of water while 
appearing dry by ordinary standards. This moisture is not held 
fast but tends to increase or decrease in amount with the rise and fall 
of the humidity in the immediate surroundings. Since film is usu- 



10.0 
9.0 
8.0 
7.0 
6.0 
5.0 
4.0 
3.0 
2.0 
1.0 



z C 
u ^ 
o * 



20 40 60 80 

PER CENT RELATIVE HUMIDITY 



FIG. 1. Relationship between relative humidity and 
moistuie absorbed by nitrate negative-type motion 
picture film starting from the dry condition. 

ally surrounded by an atmosphere which is composed of a mixture 
of air with varying proportions of water vapor, gain or loss of mois- 
ture therefore involves an exchange between it and this atmosphere, 
the moisture content of which may, in turn, be controlled by contact 
with a desiccant, by passage through an air-conditioning apparatus, 
etc. 

Moisture Content of Film. A typical motion picture negative 
film will attain the maximum of about 10 per cent moisture con- 
tent* in this way when it remains for a sufficient time in contact with 
an atmosphere which is fully saturated with moisture (100 per cent 



* Only the readily exchangeable moisture content is considered in this paper. 



478 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

relative humidity). As atmospheric relative humidity is reduced, 
the moisture content of the film decreases but not in direct propor- 
tion. The amount of moisture taken up at various relative humidi- 
ties by a typical motion picture negative film, starting from the 
dry state,* is shown for a temperature of 70 F by the curve in 
Fig. 1. (Hysteresis effects causing a small difference in moisture 
content, according to whether the film has previously been dryer 
or more moist, can be overlooked here.) The effect of temperature 
at constant relative humidity is relatively slight over the range of 
interest. Considering the uncertainties resulting from other factors 
in the usual desiccating problem, it is permissible to neglect the 
effect of temperature in this respect. 

Relative Humidity and Atmospheric Moisture Content. Rela- 
tive humidity is defined as the actual water vapor pressure in a 
given atmosphere divided by the maximum vapor pressure which 
water can exert at the prevailing temperature. Since the maximum 
vapor pressure which water can exert increases with increase in tem- 
perature, the vapor pressure' at every value of relative humidity like- 
wise increases proportionately. However, this increase represents 
only a greater availability of moisture in the space in question but not 
any tendency to increase the equilibrium moisture content of photo- 
graphic film. The quantity of moisture in grains per cubic foot of 
space is directly proportional to the existing vapor pressure and can 
be obtained from the usual charts and tables. It is assumed in this 
paper that such psychrometric charts are available to the reader. 

The presence or absence of air can be ignored in considering the 
equilibrium moisture content of film at any given temperature and 
per cent of vapor pressure saturation (relative humidity). In spite 
of its passive character as regards the equilibrium moisture content 
of the film, the air, constituting upwards of 90 per cent of the at- 
mosphere, forms an important obstacle to the movement of moisture 
by diffusion and makes up the bulk of the material which moves when 
convection takes place. The air is thus a factor tending to retard 
movement of moisture when conditions favor stagnation and, on the 
other hand, is the medium in which moisture vapor is conveyed from 
one place to another when conditions are conducive to circulation. 

In the open, local short-term fluctuations in vapor pressure are 

* Dried to substantially constant moisture content over concentrated sulfuric 
acid. 



June, 1946 USE OF DESICCANTS 

accommodated by displacement of air which is always at hand to 
maintain full barometric pressure. Within tight enclosures, such ac- 
commodation is impossible except insofar as changes in the shape of 
the container occur. Even if no accommodation is made, the greatest 
possible change in total pressure which would result (about 10 per 
cent) is not very great when it is considered that the barometric 
fluctuation is about 10 per cent and the effect of a 5000-ft ascent from 
sea level about 15 per cent. The moisture conditioning of film, 
therefore, always takes place in an atmosphere comprising a small 
proportion of moisture in a large proportion of air unless the latter is 
reduced by operating under a vacuum. 

Such are the atmospheres which serve as the medium of moisture 
transmission and in a small degree as a moisture reservoir. The small 
storage capacity of air is evident from an example. One cubic inch of 
motion picture negative film might contain about 10 grains of mois- 
ture when in equilibrium with an atmosphere containing only 0.004 
grain per cu in. (70 per cent RH at 70 F). If one cubic inch of this 
film was to be dehumidified by being sealed in an enclosure with 
sufficient air at 30 per cent RH and 70 F to provide for eventual 
equilibrium of both at 50 per cent RH, the enclosed space would 
have to be about 2900 cu in. (1.68 cu ft). 

Moisture Exchange and Equilibrium. It may be assumed that 
the moisture content of film is adjusted during manufacture to a 
value which makes it suitable for normal use. It is possible, fur- 
thermore, with modern moisture-resistant packaging, to prevent 
gain or loss of moisture until it is opened for use. If the atmosphere 
to which the film is then exposed is unfavorable in respect to humidity, 
it will be desired to handle the film in such a way as to retard change 
in moisture content and then later to hasten the return to a suitable 
moisture content. In the humid tropics, for example, the film may 
be exposed during a few minutes or a few- hours to relative humidities 
up to 90 per cent and then stored for days or weeks before develop- 
ment. In such a case excess moisture can be removed shortly after 
the film comes from the camera by enclosing it with a desiccant 
in a tight container. The desiccant should not come in direct con- 
tact with the film. 

Before considering the choice of desiccants and operating methods 
it will be desirable to consider some of the factors which govern the 
rate of exchange of moisture between the film and its surroundings as 
follows : 



480 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

(1) The humidification or dehumidification potential, 

(2) The length and cross section of diffusion paths, t 

(3) Circulation, 

(4) Temperature. 

. - 

By implication are included such factors as the disposition of the 
film, that is, in a roll or stack of sheets, etc., the character and ar- 
rangement of wrappings or other membrane, the location of desiccant, 
and limitations of moisture capacity of elements in the system. 

Humidification Potential. Moisture equilibrium between a sub- 
stance and its surroundings is reached when the moisture vapor 
pressure which tends to exist inside the substance equals that pre- 
vailing in the system surrounding it. Only when a vapor pressure 
difference exists does moisture exchange take place. This is so in the 
case described when the film absorbs moisture from a humid outdoor 
atmosphere or later when the film and desiccant within an enclosure 
exchange moisture through an intermediary atmosphere, the rela- 
tively small moisture content of which is controlled by them. Dif- 
ference in vapor pressure in 2 substances, whatever the cause, is a 
measure of the tendency of moisture to transfer between them and 
will be termed the humidification or dehumidification potential. 
It should be recognized that the moisture vapor pressure in a sub- 
stance such as photographic film increases with increase in tempera- 
ture and also with moisture content but not in a simple fashion. 

Length and Cross Section of Diffusion Paths. At one or more 
points along the path of moisture exchange, the mode of transport 
is one of diffusion. In diffusion processes the rate of transfer at any 
point in the path is directly proportional to the concentration gradient 
and the cross-sectional area. If the humidification potential is fixed, 
the rate will decrease with increase in length of the diffusion path. 

In the case of desiccation^ the rate is therefore increased by spread- 
ing out the film for access, using a desiccant of such open structure as 
to introduce minimum constriction of the diffusion path, placing the 
desiccant close to the film (but not in contact!) and eliminating mois- 
ture-impervious wrappings, or if mechanical membranes are indispen- 
sable, choosing the most pervious. 

If humidification of film is to be retarded without the -aid of desic- 
cants, the measures to be adopted will generally be the opposite of 
those listed. 

Diffusion through air is slow so that transfer is accelerated by re- 



June, 1946 USE OF DESICCANTS 481 

moving any kind of obstruction to circulation which forms paths 
through which transfer must be by diffusion. Since the presence of 
air provides resistance to moisture diffusion, the process is speeded up 
by pumping the air out of the enclosure in question. 

Diffusion through the thickness of the film itself is also slow so that 
when the other limiting factors are largely eliminated (under vacuum), 
it still requires about an hour to effect a large change in the moisture 
content. On the other hand, such a change may require up to several 
days or weeks if the other factors have full play. 

Experimental results to be described in a later section will illus- 
trate the magnitude of some of the effects mentioned here. 

Circulation. In contrast to the slow movement of moisture by 
diffusion through stagnant air, the effect of any noticeable degree of 
air movement is large. A really stagnant condition prevails less 
frequently than might be supposed except where space is cut up into 
small cells by solid material. In spaces having dimensions of a few 
inches, convection may be considerable as a result of heating and cool- 
ing, or differences in atmospheric density in accordance with differ- 
ence in moisture content. Most rapid moisture transfer occurs not 
simply when circulation is maintained in the container as a whole but 
when the air stream moves rapidly over all the surfaces of film, desic- 
cant, or other bodies where moisture effusion or extraction occurs. 
Moisture transfer is therefore retarded greatly by the mere presence 
of wrapping material so far as it obstructs circulation, and even more 
so by wrappings of moisture-impervious material which permit ex- 
change only by diffusion along the folds and between layers. 

Temperature. Temperature effects are of several kinds. In the 
first place, diffusion rates increase with temperature. Rise and fall 
of temperature in the surroundings produce convection in small ves- 
sels, especially if heating and cooling are rapid and localized. At 
higher temperatures, the atmosphere can contain more moisture 
and hence is a better transfer medium. The decrease in moisture 
content of film with elevation of atmospheric temperature at con- 
stant relative humidity is not large enough to be of much practical 
significance in the management of film moisture content. Most des- 
iccants lose their moisture absorption capacity with .elevation of 
temperature but usually not so much over the range of interest as to 
affect the choice of agent or the manner of use. 

Practices in which humidification or desiccation occur as a result 
of change of relative humidity brought about by heating or cooling of 



482 



C. J. KUNZ AND C. E. IVES 



Vol 46, No. 6 



air, including those cases where moisture is extracted as dew or frost, 
will be discussed in a later section. 

Combination of Several Factors. Since diffusion of moisture 
through the film itself is slow, movement of moisture to and from 
the inner layers of film in wound rolls or in stacks of sheets is affected 
greatly by the tightness of winding or packing. In a very loosely 
wound roll the rate may be one tenth of that for an open sheet and in a 
tight roll even less. Similar effects must be expected with sheet films 
although the closeness of adjacent sheets will be a matter of consider- 
able uncertainty. Another consequence is that the outer edges will 





UJ U- 
ffi O 



O uj 

u a. 



200 



120 



80 



40 



14 



16 



rO-^CJ>-gCDOOU3 < tO U> O 
000 00 01 Wi CO tO 



18 



ENT RELATIV 
EQUILIBRIUM 



2 46 8 10 12 

TIME (DAYS) 

FIG. 2. Absorption of moisture by a loosely wound roll of nitrate negative 
motion picture film, circulation not forced. 



experience more rapid humidification and dehumidification than the 
inner portions. 

The effect of the presence of metal foil and waxed paper wrappings 
as well as magazines, metal film-pack cases, etc., is to create long nar- 
row diffusion paths which so reduce the rate of transfer as to offer 
substantial protection from humidification if used properly and, on the 
other hand, to demand special procedures in desiccation. 

For like reasons, the desiccating chamber must provide for ex- 
posing a large area of desiccant and avoiding obstruction of circula- 
tion by the contents. 

As will be shown in a later section, the rate of moisture removal 
may be reduced if the desiccant is not in excess. Limitations in the 



June, 1946 USE OF DESICCANTS 483 

capacity of the system must also be respected when the space for 
desiccant is small or the desiccant is bulky or low in capacity. 

Rate of Moisture Absorption. An open strand or sheet of film 
in air which is moving even at a very slow rate can absorb moisture 
from a humid atmosphere at such a rate as to effect a large propor- 
tion of the total adjustment within about an hour and almost all of 
the adjustment in 3 or 4 hours. The rate of absorption is reduced 
greatly by the presence of any kind of barriers to circulation and 
diffusion. Fig. 2 shows the slow progress of moisture absorption 
by motion picture film in a roll, starting from the bone-dry state and 
exposing the roll to high humidity under conditions where circula- 
tion was very little. The effect of tightness of winding is demon- 
strated by the data in Table 1, which shows the progress of moisture 



TABLE i 

Effect of Tightness of Winding on the Rate of Absorption of Moisture with Nitrate 
Negative Motion Picture Film 

Weight of Moisture 

Absorbed by 1000 Ft Time Required to Absorb a Given Weight of Moisture (Min) 

of Film (Gm) Tight Loose Thread 

50 66 52 27 

60 117 90 42 

70 180 132 63 

80 258 177 87 

90 345 231 117 

100 468 303 153 

110 648 400 216 

120 840 540 285 



absorption in 3 rolls of film, one of which was tightly wound, the 
second loosely, and the third with an interwinding of No. 8 cotton 
thread for spacing. The film was originally of very low moisture 
content and during the test was held over water in a jar of ample 
dimensions to permit convective effects. A rough estimate of the 
time required for a change of intermediate degree can be made from 
this. Greater degrees of retardation are obviously possible by the 
use of tightly wound and closely folded wrappings of metal' foil, heavy 
waxed paper, and the like. Where wrappings of this kind can be 
used and especially if film holders can be dried before loading, it 
should be possible to reduce to a very small amount the absorption 
of moisture when the film is out of the camera. 



484 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 



DEHYDRATION METHODS 

Having considered the factors which influence absorption and re- 
moval of moisture and the harm which excess moisture can do, we 
shall now turn our attention to detailed consideration of methods 
for removing moisture as they may be applied to such material. 

In general, dehydration of a space containing air and a solid body 
can be accomplished by any one of, or a combination of, certain of 
the following processes : 

(1) Condensation of water vapor by refrigeration and subsequent reheating of 
the air, 

(2) Drying at reduced atmospheric pressure, 

(5) Desiccation involving chemical action (absorption), 
(4) Desiccation involving physical attraction (adsorption). 

Although seemingly somewhat outside the scope of this paper, 
the first 2 methods listed require consideration since they offer ad- 
vantages in some film dehydration problems. 

Dehydration by Refrigeration and by Vacuum General Prin- 
ciples.; All methods of dehydration involve reduction of the mois- 
ture vapor pressure in the space surrounding the object to be treated, 
below that prevailing in that object. 

In the case of refrigeration, the vapor pressure of the space is de- 
creased by reducing the temperature in a portion of that space to the 
dew point or lower ; the method of dehydration employing a vacuum 
depends on the action of a pump to reduce the pressure within the 
given space either to the extent of removing the moisture mechani- 
cally or of facilitating its removal by desiccants. 

Application of Refrigeration to the Dehydration of Film. The 
following comments can be made with reference to dehydrating 
film by the method of refrigeration : 

(1) A system of refrigeration employing melting ice is unsuited for the de- 
hydration of film because it tends to keep the atmosphere near saturation with 
moisture so that humidification is more likely to occur. This is demonstrated 
by the fact that fruits and vegetables, stored in an ice refrigerator, seldom dry out 
but remain firm and crisp. 

(2) Mechanically operated refrigerating systems employing a cooling medium 
which circulates through suitably disposed piping is capable of dehydrating film 
only when the vapor pressure of the latter exceeds that existing at the cold surface. 
Since vapor pressure is dependent on temperature, the problem is one of maintain- 
ing the temperature of the film well above that of the cold surface. Because of 
the nature of the sensitized film, the methods for accomplishing this are restricted. 
One method consists in circulating air within the refrigerating chamber in such a 
manner that, after passing over the cooling surface and giving up its moisture, 



June, 1946 USE OF DESICCANTS 485 

the air is warmed before continuing on to the film whether another load of moisture 
is taken up. 

Following dehydration to the desired moisture content, the film should be 
wrapped in black paper, sealed in a metal container, and stored until ready for 
processing, preferably under refrigerated conditions. 

Effect of Refrigeration on the Relative Humidity within a Sealed 
Film Container. It is appropriate to consider the case in which 
film, in equilibrium with an atmosphere of, say, 60 per cent RH at 
70 F, is sealed in a container and- placed in refrigerated storage. 
The question arises, will the relative humidity within the sealed con- 
tainer increase as the temperature is lowered, with the attendant 
danger of condensation of moisture? An analysis of existing condi- 
tions shows that the relative humidity within a sealed container of 
normal packing size will remain substantially constant over any 
temperature range which will ordinarily be encountered. For in- 
stance, if the film package contained no film, the enclosed space 
would, in fact, be saturated at 55 F since this is the dew point of the 60 
per cent, 70 F atmosphere. However, when film and wrappings are 
present and sufficiently accessible to the air in the container, moisture 
is removed from the space by the film and the paper as the relative 
humidity begins to rise, as a result of the lowering of temperature. 
Obviously, if the moisture in this space is thus reduced in quantity, 
saturation will not occur at 55 F. Since the moisture capacity of 
the film and paper is great compared with that of the surrounding 
space, a substantial reduction in temperature changes the existing 
relative humidity only a very small amount. In this adjustment the 
change in moisture content of the film is relatively minute. 

Dehydration by Vacuum. The utility of the method of dehy- 
drating in a vacuum is demonstrated when one considers that the 
rapid dehydration of many of the food products for overseas ship- 
ment during the war was accomplished by this method applied in 
one form or another. 

When applied to the partial dehydration of overhumidified film, 
the method is attractive in many respects but primarily because it is 
rapid. Since the extent of latent-image impairment is a function 
of the time of exposure to unfavorable conditions, acceleration of the 
dehumidification process is desirable. Also, since the operation 
may be reduced in time from days to hours, a smaller amount of 
drying space is needed. 

The necessary equipment need not be bulky in itself, since the es- 



486 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

sential parts consist merely of a pump for creating the vacuum, and a 
dehydrating chamber of a size consistent with the quantity and type 
of film to be treated. 

The simplest method of utilizing a vacuum for the purpose is 
merely to evacuate the film chamber to a pressure substantially 
less than the existing vapor pressure of the film. As soon as vapori- 
zation commences, the temperature of the film will fall for reasons 
already given so that the initial rate of dehydration will decrease 
somewhat. However, the final temperature attained after starting 
from 70 to 100 F is still such as to permit sufficient dehydration in 5 
to 10 hr, provided a low pressure (one millimeter of mercury or less) 
(one millimeter = 0.04 in.) is maintained. This vacuum method is 
effective with firmly wound rolls of motion picture film and is the 
method preferred in all cases where the disposition of the film is simi- 
larly unfavorable. 

Although the use of a trap for water vapor, i. e., a refrigerated con- 
denser, between the pump and the desiccating chamber permits the 
use of a pump of much lower capacity, such a system would not ap- 
pear desirable for field application, because what might be gained in 
the size of the pump, would be sacrificed in refrigeration equipment. 

Another method for utilizing a vacuum for dehydrating film con- 
sists in decreasing the pressure to a value which is equal to or even 
somewhat greater than the vapor pressure of the moisture in the film 
while providing a suitably disposed moisture-absorbing body for 
instance, a desiccant or refrigerated surface condenser. In this case 
the effect is merely that of reducing the resistance to the passage of 
water vapor from the film to the absorbent or the extracting surface, 
by removing most of the interfering air molecules. Practical as- 
pects of this subject are considered further in a later section. 

Dehydration by the Use of Desiccants. We now return to the 
consideration of the remaining methods of dehydration, broadly 
classified as methods employing desiccants. In simplest terms, a 
desiccant may be described as any agent having a greater affinity for 
absorbed moisture than the substance to be dehydrated so that 
water vapor passes from the latter to the former until a state of mois- 
ture equilibrium is attained. 

Properties of a Desiccant Suitable for Dehydration of Film. 
The properties of a desiccant which is suitable for the dehydration of 
film are not, in every case, the same as those required for other ap- 
plications. In fact, even with film, properties which are essential 



June, 1946 USE OF DESICCANTS 487 

when the desiccant is to be used in one way are unimportant when it is 
employed in another. However, in general, the following properties 
would normally be considered desirable in a desiccant which is to be 
used in the partial dehydration of film, especially when used in the 
field: 

(1) The moisture absorption capacity of the desiccant should be large. 

(2) The velocity of absorption should be high, at least down to the level of 
relative humidity considered suitable for safe film storage. 

(5) The specific volume should be low for minimum space requirements. 

(4) The physical form of the desiccant should not change in an undesirable 
manner, even when fully saturated. As an example, liquefication or expansion 
should not occur. Freedom from the tendency to form dust or powder is desir- 
able. 

(5) The effect of temperature on absorption velocity and capacity should not 
be great over the. range of working temperatures. 

(6) For some applications, the desiccant should be capable of regeneration 
without deteriorating. 

(7) The desiccant should not, in itself, produce undesirable changes in the 
sensitometric properties of photographic film. For instance, certain substances 
which might conceivably be used as desiccants have been shown to produce vapors 
which have considerable effect on the latent image. In general, neutral drying 
agents are to be preferred to those which are acidic or basic. 

(8) The substance should be essentially inert, especially in the sense that it 
should be incapable of inducing or supporting combustion. This is of special 
importance, considering the combustible nature of film. The very effective per- 
chlorate desiccating agents are examples of a type to be avoided for this reason. 

It may be recognized that one of the properties ordinarily associ- 
ated with a good desiccant "drying efficiency" is lacking from the 
list. In the literature of desiccants, this is a measure of the ability 
of an agent to dehydrate to a condition approaching complete dry- 
ness. Since this factor is of little importance in the present applica- 
tion where a vapor pressure corresponding to a relative humidity 
of 20 or 40 per cent is all that is required, certain desiccants of wide 
reputation fall into a class of only academic interest. For instance, 
phosphorous pentoxide is often referred to as a standard of drying ef- 
ficiency in view of its ability to produce a vapor pressure of prac- 
tically zero ; however, it has a very small capacity for moisture less 
than 10 per cent because of the formation of gummy metaphos- 
phoric acid on its surface, which effectively seals the agent against 
further absorption, at least at reasonable rates. Barium oxide is an- 
other exceedingly powerful desiccant of very low capacity. 

Factors Affecting the Action of a Desiccant. Before proceeding to 
consider particular agents and methods of use, we shall discuss 



488 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

briefly certain factors which determine their suitability according to 
the type of use. 

As already stated, desiccants can be divided into 2 general groups, 
one depending on chemical action (absorption) and the other on 
physical action (adsorption). Except where classification in this 
regard is discussed, the general term "absorption" is used inclusively. 

The first of these 2 groups can be divided into 2 subgroups ; in one 
of these, the desiccant absorbs moisture by virtue of a shift in pri- 
mary valences, thus creating another molecular species; in the sec- 
ond subgroup, the desiccant absorbs by virtue of hydration, thus 
involving only the use of secondary bonds or valences. 

Examples of each of these types of desiccant are shown in Table 2 : 

TABLE 2 

Examples of Desiccants Involving (a) Absorption and (&) Adsorption 

(A) Absorption (Chemical) 

(a) Reaction in the usual sense 
P 2 O 5 + 3H 2 O > 2H 3 PO 4 
BaO + H 2 O > Ba(OH) 2 
(6) Hydration 

CaCl 2 + OOHoO - CaCl 2 - (x)H 2 O 
CB) Adsorption (Physical) 
(a) Silica Gel 
Charcoal 

The highest drying efficiency is possessed by agents which owe 
their effectiveness to a chemical action involving molecular rearrange- 
ment. The temperature coefficient of decrease in dehydrating ef- 
ficiency with increasing temperature is lowest with this type of agent. 
In general, desiccating agents of this type are not readily regenerated. 
Unless some adsorptive effect occurs simultaneously, the capacity of 
these agents is ordinarily low, being limited to that represented in 
familiar chemical reaction equations. Barium oxide is an agent of 
this type which owes part of its limited capacity to adsorption. 

The drying efficiency of agents relying on the formation of hy- 
drates depends on the vapor pressure of the hydrate formed. The 
vapor pressure capable of being attained by this method varies from 
practically zero with, for example, anhydrous magnesium perchlorate, 
upward through 0.14 mm mercury at room temperature with granular 
calcium chloride. The hydration- type reaction results in greater 
moisture capacity since the number of molecules of hydration is usu- 
ally large. Moreover, adsorption effects are often associated with 



June, 1946 USE OF DESICCANTS 489 

chemical hydration, increasing the capacity and rate of absqrption 
of this type agent considerably. The temperature coefficient of 
change in dehydrating efficiency is somewhat more than when molecu- 
lar rearrangement is involved. Many of the agents included in this 
group are capable of being regenerated, especially if only a small to 
.moderate amount of water has been absorbed. However, some agents 
tend to melt or fuse at the elevated temperatures required for re- 
generation at normal pressure, thereby altering their physical form 
so that they no longer have favorable pore structure or particle size. 

The second general group is made up of those agents which de- 
pend for their action merely on adsorption. These agents consist of 
materials which, by their nature or manner of preparation, have a 
porous structure consisting of capillary openings which are in size 
often beyond the resolving power of optical microscopes. Condensa- 
tion of moisture in these capillaries results in a very great diminution 
in the vapor pressure of the material adsorbed. The drying efficiency 
of many of these agents (for very small amounts of moisture ab- 
sorbed) can be shown to be high, in some cases being equal to or 
greater than, for instance, that of concentrated sulfuric acid. The 
capacity for moisture of such substances varies from comparatively, 
small (18 per cent with activated alumina A^Os) to very substan- 
tial amounts (45 per cent with silica gel). In becoming saturated, 
the physical appearance of most of these substances remains un- 
altered. As a class they can be regenerated readily by heating to a 
temperature somewhat over 250 F. 

Desiccants Suitable for Dehydration of Film. In view of the 
special requirements of a desiccant to be used for the partial de- 
hydration of film, the choice of suitable agents is considerably re- 
stricted. The following list, however, includes some of the agents 
which are either (1) generally well adapted for the purpose, or which 
have outstanding individual properties, or (2) which have been 
recommended in the past for the purpose of dehydrating film : 

Calcium Chloride 

Calcium Sulfate (Drierite, Anhydrite) 

Silica Gel 

Aluminum Oxide (Activated Alumina) 

Tea 

Rice 

Paper 

Absorbent Cotton 

Charcoal 



490 



C. J. KUNZ AND C. E. IVES 



Vol 46, No. 6 



Calcium chloride and calcium sulfate are widely used desiccants 
of the hydrate-forming type having some adsorptive properties, 
while silica gel and activated alumina are representative of the purely 
adsorptive type. Tea, rice, paper, absorbent cotton, and charcoal 
are substances having some absorptive capacity which have been 
mentioned in the past for use in dehydrating film. 

Absorption of Moisture by Typical Useful Desiccants. The 
percentage increase in weight of certain of these desiccants over 



100% 




4 6 10 
DAYS 



TIME 



FIG. 3. Increase in weight of various desiccants at high relative 
humidity, circulation not forced. 



their initial dry weight, under conditions of high (approximately 100 
per cent) relative humidity, has been plotted against time in Fig. 3. 
In making these measurements, the weight of a sample of desiccant 
(initially 10 grams) enclosed in a small desiccator (20 cu in.) contain- 
ing distilled water at 70 F was taken at intervals until a constant 
weight was indicated. These measurements therefore are based on a 
static Condition, such as might be encountered in a closed chamber 
containing only film and a desiccant. " 

Fig. 3 also shows what we have chosen to term the "ultimate mois- 
ture content" of these desiccants, that is, the maximum weight of 



June, 1946 



USE OF DESICCANTS 



491 



moisture which a desiccant can absorb at a stated relative humidity 
and temperature. 

It is seen that under these conditions the ultimate moisture con- 
tent of calcium chloride exceeds a 100 per cent increase in weight of 
the dry desiccant. Because this particular agent liquefied after ab- 
sorbing about 60 per cent of its dry weight in moisture, measurements 
were discontinued when it had increased in weight by 100 per cent; 
therefore the ultimate moisture content is not shown. 



100% 



10% 




TIME 



FIG. 4. Increase in weight of various desiccants at low 
relative humidity, circulation not forced. 



With silica gel, the weight increases in much the same manner as 
with calcium chloride but reaches a maximum after increasing by 
about 45 per cent. 

The ultimate moisture content of the rest of the agents included 
in these measurements, for practical purposes lies between a 10 to 
20 per cent increase in weight, though the time required to reach this 
condition varies with the different desiccants. 

Other samples of these agents were brought from the dry state to 
equilibrium with an atmosphere of about 30 per cent relative humid- 
ity by the use of a suitably adjusted solution of sulfuric acid in the 
desiccator. Fig. 4, plotted to the same scale as Fig. 3, shows the 
course of absorption under these conditions. The decrease in ulti- 



492 



C. J. KUNZ AND C. E. IVES 



Vol 46, No. 6 



mate moisture content at the lower value of relative humidity is 
evident by comparison of Figs. 3 and 4. 

Effect of Circulation on Rate of Absorption and on Ultimate 
Moisture Content. Another of the factors referred to earlier as 
affecting the rate of exchange of moisture between film and its sur- 
roundings was circulation. The same effect is shown, of course, when 
a desiccant is considered instead of film. This is illustrated by com- 
parison of Figs. 3 and 5. The conditions existing in collecting data 
for Fig. 5 were basically similar to those described in relation to Fig. 3 ; 
however, a small fan circulated the air within the desiccator in the 



100% 



10% 




7 



100 



CALCIUM CHLORIDE 



1000 MINUTES 



I 2 3 4 5 6 8 12 20 40 HOURS 

TIME 

FIG. 5. Increase in weight of various desiccante at high relative 
humidity, forced circulation. 



case of Fig. 5. It will be noted that the rate of absorption was in- 
creased 8 to 10 times by the use of circulation. (Note that the 
time scale in terms of "days" in Fig. 3 compares with "hours" in Fig. 
5.) - 

It will also be noted that the ultimate moisture content of these 
desiccating agents remained substantially unchanged, showing that, 
if given sufficient time, the same ultimate moisture content can be 
attained, whether or not circulation is used. 

Ultimate Moisture Content of Typical Desiccants over Whole 
Range of Relative Humidity. The increase in weight of some of 
these desiccants, resulting from the absorption of moisture at 
various levels of relative humidity (ultimate moisture content) is 



June, 1946 



USE OF DESICCANTS 



493 



shown in Fig. 6. These measurements were made at 70 F. While an 
increase in temperature would decrease the moisture content, at a 
stated value of relative humidity and a decrease in temperature would 
increase the moisture content, the effect in no case would exceed =*= 5 
per cent of the values given for a temperature variation of =*= 20 F. 

Disposition of Desiccant with Respect to the Film. Another 
factor affecting the rate at which moisture transfers from film to 




20 40 60 80 100 

PERCENT -RELATIVE HUMIDITY 
* SURFACE MOIST IN THIS RANGE. MOISTURE CAPACITY NOT WELL DEFINED 

FIG. 6. Increase in weight of desiccants resulting 
from the absorption of water vapor at various levels of 
relative humidity. Increase in weight referred to weight 
of sample conditioned at zero per cent relative humidity, 
temperature, 70 F. Curve (a) plotted from data pub- 
lished by manufacturer; Curve (6) plotted from data 
obtained in the present work; Curve (c) same as (b), 
averaged results. 



desiccant is the size and shape of the desiccating chamber or of the 
cells or subdivisions within such a container. While it is true that 
moisture will eventually diffuse from one section to another as long 
as the parts are connected, the rate at which such diffusion occurs is 
highly dependent on the relative proportions. This was shown in a 
simple experiment in which moisture was allowed to diffuse upward 
through a glass tube, one inch in diameter, 8 in. long, and closed at 
the top. The tube was supported above a small vessel of water at 



494 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

room temperature (70 F). Punchings from a blotter which pre- 
viously had been soaked in a solution of cobalt chloride and then 
dried, were arranged at various levels in the tube, so that a color 
change from blue to pink indicated that a relative humidity of at 
least 50 per cent had been attained at a particular level. These 
tests showed that, at constant temperature, at least 24 hr were re- 
quired to fill the tube with water vapor. Similarly, when concen- 
trated sulfuric acid was substituted for water, the same period of 
time was required to dehydrate the entire tube, from top to bottom. 

While this test showed that moisture transfer takes place some- 
what slowly through a long channel or tube, another similar experi- 
ment showed that it occurs quite rapidly in a more favorably propor- 
tioned chamber. Small cobalt-chloride impregnated punchings, ar- 
ranged at various levels in a squat laboratory-type glass desiccator 
jar all changed color within an hour or two, whether the process was 
one of desiccation or humidification. 

TABLE 3 

Desiccating A gents A rranged in Order of Increasing Specific Volume 

Specific Volume 
(Cu In. Per Lb) 

(1} Calcium Sulfate 29 

(2) Black Photo Wrapping Paper Arranged Sheet upon Sheet 32 * 

(5) Rice 32* 

(4) Activated Alumina (14-20 mesh) 32 

(5) Silica Gel (8-20 mesh) 43 

(6) Calcium Chloride 44 

(7) Tea 60* 

(8) Charcoal 

(9) Crumpled Black Photo Wrapping Paper 280 or more 

* Approximate, depending on sample and treatment. 



These 2 tests have been cited to illustrate the type of condition to 
be avoided, in one instance, and to be sought in another. Practically, 
it might be pointed out that the size and shape of wound rolls of mo- 
tion picture film tend to dictate that a desiccant (i)>e arranged in the 
form of a ring with the film roll resting in the center or, better, (2) be 
spread out in a layer above and below the roll of film which lies flat in 
its container. Either scheme distributes the desiccant with respect 
to the film and provides a short path. 

With some film products other than motion picture film, their size 



June, 1946 USE OF DESICCANTS 495 

and shape are such that they could be packaged to form a long con- 
stricted channel between the more remote parts and the desiccant. 
Such a condition is, of course, to be avoided. 

Volume- Weight-Packing Relationships. One of the factors af- 
fecting the space in a film desiccating container which must be 
devoted to the desiccant is the specific volume of that agent that is, 
the volume occupied by a unit weight of the substance in question. 
Table 3 is made up of those agents listed as typical of more satisfac- 
tory agents, and is arranged in order of increasing specific volume 
the space required by agents farther down in the list is greater than 
that required by those above. 

Closely allied to the specific volume of the bulk material is the man- 
ner in which the desiccant tends to pack in its container. Coarse, 
granular particles, such as silica gel,* rice, alumina,* calcium chlo- 
ride,* or calcium sulfate,* pack'in such a manner that connected voids 
exist between particles which permit the passage of moisture, thus 
promoting a higher rate of desiccation. The internal structure of the 
particles is important in this same respect. Tea leaves tend to pack 
with smaller connecting voids while stacked paper offers a very poor 
circulatory path. Crumpling the paper improves this situation 
but the specific volume increases greatly. The use of paper confetti 
would improve the circulatory path somewhat without increasing the 
bulk to the extent caused by crumpling the paper. 

Membrane Separating Film from Desiccant. The last of the 
factors to be discussed, which influences the rate of diffusion of 
moisture from film to desiccant, is the effect of any film wrapping or 
membrane separating the 2 materials. In any case where the film de- 
humidifying chamber is to be moved about during use, it is necessary 
to shield the film from even small particles of the desiccant. Such 
particles are likely to cause mechanical damage or spots on the de- 
veloped image. Sheet or membranelike materials which will trans- 
mit water vapor but filter out the dust offer an effective means for 
separating the 2 materials. In some cases they can be used to wrap 
the film, provided all openings are cemented shut; in others, they 
may be used to confine the desiccant in a chamber of its own. 

While the use of such moisture pervious membranes may be neces- 
sary under some circumstances, they all have a restraining effect on 
diffusion. This is evident from Table 4, in which the rate of diffusion 

* In the forms supplied for drying. 



496 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

of water vapor through various materials is compared with the rate 
existing when no membrane was present. 

The test membrane was cemented over the opening in a small 
wide-mouth jar in which was an ample supply of desiccant. Moist 
air was circulated over all of the jars by means of a desk fan to avoid 
the effects of local depletion, uncontrolled convection, etc. 

TABLE 4 

Rate of Transfer of Moisture by Membranes under Effect of Large Humidifying 

Potential 

Relative Rate of Transfer 
Membrane Material of Moisture (Per Cent) 

(1) No membrane 100 

(2) Filter Paper (Will Corp. No. 7679) 25 
(5) Matted Fiberglas pad (Vie in. thick) 20 
(4} Chamois ("Photo Chamois" E. K. Co.) 20 

(5) Thin interleaving paper (black photo) 18 

(6) Blotter (Photo-Blotter, E. K. Co.) 16 

(7) Film Pack Paper (E. K. Co.) 16 

(8} Film Wrapping Envelopes (various weights) 13 to 8 

It is evident from these results that the comparatively heavy 
papers used for film envelopes and wrappings are among the more 
impermeable to the passage of moisture. The Fiberglas pad, chamois, 
filter paper, and interleaving paper make up a group of more perme- 
able materials which offer a minimum of resistance. The filter paper 
and lightweight paper may suffer from the disadvantage of becoming 
somewhat brittle. 

Methods of Utilizing Desiccants in Dehydrating Film. Up to 
this point the discussion has involved (a) the factors affecting the ab- 
sorption of moisture by film, and (b) the factors applying to the 
choice of desiccant. At this time, therefore, we turn to considera- 
tion of the use of suitable desiccants in the partial dehydration of 
film. 

Two general methods of utilizing desiccants for removing moisture 
from film can be used. The first, capable of the more rapid action, 
consists in holding the film temporarily in a suitable container with a 
large excess of a powerful desiccant. When sufficient time has 
elapsed to dry the film to a proper moisture content it is removed and 
sealed in another container without desiccant until processing is 
undertaken. This procedure will be referred to in the future as the 
"temporary desiccator" method. 



June, 1946 



USE OF DESICCANTS 



497 



The second method, which should only be considered when the 
need for rapidity in desiccating is less, involves packing the film 
immediately after use with a quantity of desiccant estimated ac- 
cording to the particular situation to provide for eventual equilibrium 
at a selected value of relative humidity. The film container is not 
reopened until the time of processing. This will be referred to as the 
"permanent packaging" method. 




DAYS 

FIG. 7. Desiccation of a loosely wound roll of nitrate 
negative motion picture film, circulation not forced. 
Film initially in equilibrium with 93 per cent relative 
humidity temperature 70 F. Curve (1) film desiccated 
over large excess of concentrated sulfuric acid, normal 
pressure; Curve (2) film desiccated over a limited 
amount of silica gel, normal pressure; Curve (5) same 
as (1) but at reduced pressure (2 in. mercury, absolute). 



A variation of the second method seems to be the one most often 
recommended in the past when a film desiccating problem arose. 
However, in many cases the benefit which could be gained from the 
quantities of such low-capacity desiccants as dried tea, rice, or paper 
sometimes suggested could hardly have been noticeable. This will 
become more apparent as we consider the quantitative factors con- 
cerned. 



498 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

To illustrate the operation of these 2 methods of utilizing desic- 
cants, the following experimental evidence may be cited. 

Laboratory Test Involving Temporary Desiccator Method. 
In an experiment involving the temporary desiccator method, 
a loosely (but not openly) wound roll of motion picture film (nitrate 
negative) in initial moisture equilibrium with an atmosphere of 
about 93 per cent relative humidity was suspended in a desiccator 
over a large excess of a powerful desiccant in this case, concentrated 
sulfuric acid. The film was arranged in such a manner that its 
weight could be measured without opening the desiccator. A room 
temperature of 70 F was maintained. Circulation was not forced 
in any way but the size and proportions of the container were favor- 
able for spontaneous convection. 

The results of this test are shown in Fig. 7, Curve 1. It is seen 
that under the conditions used, the moisture content of such a wound 
roll can be reduced from high to moderate values quite rapidly, 2 l / 2 
days being required to reduce the equilibrium condition of the film 
from 93 to 40 per cent RH. At values less than 40 per cent, desic- 
cation proceeds at a rapidly decreasing rate. 

Laboratory Test Involving Permanent Packaging Method. 
Another test involving the permanent packaging method was 
carried on under conditions identical with the test just described ex- 
cept that the film and the desiccant, in this case silica gel, were of 
such quantities relative to each other as to come to equilibrium at 
about 40 per cent RH. The results of this experiment are shown in 
Fig. 7, Curve 2. It will be observed that, instead of 2 J /2 days, nearly 
10 days were required to reach the desired relative humidity. 

Temporary Desiccator Method at Reduced Pressure, A third 
test illustrates the increase in rate of desiccation which can be at- 
tained by the use of a moderate vacuum. Conditions during this 
test were identical with those described in connection with the 
temporary desiccator method except that the pressure in the vessel 
was reduced to about 2 in. of mercury. Such a pressure can often be 
attained by using a water aspirator pump. Curve 3, Fig. 7, when 
compared with Curve 1, illustrates the increase in rate of desicca- 
tion which can be obtained simply by reducing the resistance offered 
by the air to the diffusion of water vapor. Equilibrium with an an- 
mosphere of 40 per cent RH was attained in about 18 hr, compared 
with-2V2 days at normal pressure. 

Relative Usefulness of Typical Desiccants. It now becomes 



June, 1946 USE OF DESICCANTS 499 

evident that, of those requirements listed in the section "Properties 
of a Desiccant Suitable for Dehydration of Film," the principal ones 
are ' 'high moisture-absorbing capacity' ' and ' 'rapid rate of absorption. ' ' 
These are true whether the desiccant be packed with the film in the 
temporary or the permanent manner. Another requirement apply- 
ing only to the permanent method is that the desiccant should not 
liquefy. Considering these major requirements, we can now evalu- 
ate the usefulness of those desiccants selected as typical. 

Silica gel appears to be the most generally satisfactory of the 
desiccants considered for the purpose. Its capacity for moisture 
and rate of absorption are good, and it retains its original physical 
form, even on becoming fully saturated. It is inert chemically and 
presents no serious problem in packaging or handling except the 
usual one of confining the dust. ' If necessary, it can be regenerated, 
even under field conditions. 

Calcium chloride has a great moisture-absorbing capacity and a 
high rate of absorption so that in cases where its tendency to liquefy 
is unimportant, as in the case of the temporary desiccator method, 
it can be used to advantage. It must be kept in mind, however, 
that calcium chloride is an expendable material since, after liquefica- 
tion, the steps necessary to return it to useful physical form are 
generally impractical for field application. 

The low to moderate moisture-absorbing capacity and rate of 
absorption of activated alumina and. calcium sulfate, especially in the 
intermediate range of relative humidity, reduces the usefulness of 
these agents in the present application in spite of the fact that they 
can be regenerated easily. In cases where added weight and bulk 
are not important considerations, these agents would be satisfactory 
otherwise. 

Readily available materials, such as tea, rice, paper, etc., are gener- 
ally deficient in both capacity and rate of absorption. To obtain 
useful results, large quantities (volumes) are ordinarily required. 
The use of such materials will be considered, however, in the next 
section, in view of the possibility that, under some circumstances, 
they may be the only materials available for the dehydration of film. 

Besides desiccants belonging to the groups listed, many other 
agents might, under certain circumstances, be used for the purpose. 
In general, these can be classed as agents which have satisfactory 
desiccating properties, but which are unsuited for field use because 
of other undesirable properties. For instance, concentrated sul- 



500 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

furic aid is an excellent desiccant but cannot be recommended for 
general field use because of its corrosive nature. However, should 
such an agent be the only one available, it can be used if suitable pre- 
cautions are taken. Sodium or potassium hydroxide and quick- 
lime are other examples of this type of agent. 

RECOMMENDED PRACTICE 

The best advice which can be given when film is to be exposed to 
humid atmospheres and heat is: "Do everything possible to pre- 
vent the film from absorbing excess moisture. If conditions are 
unfavorable, avoid long storage of film." Then, if in spite of every 
precaution, the preservation of the latent image is not assured, an 
adequate dehydrating procedure should be applied immediately. 
It should be kept in mind that, while suitable dehydration can re- 
tard the progress of change in sensitometric properties, it can do 
nothing to correct changes which have already occurred. 

Film packaged by the manufacturer in tightly sealed sheet metal or 
flexible metal foil containers may be considered safe from change in 
moisture content at least while the film remains in the package. Such 
a film container offers no protection from the effects of high tempera- 
ture, so that it should be stored in as cool a place as possible. How- 
ever, if this is done, the film should be allowed to warm up to the 
temperature of the atmosphere in which it is to be used for a period of 
about a day; otherwise, moisture from the humid atmosphere is 
likely to condense on the film as soon as the package is opened. 

The film container should not be unsealed before it is necessary to 
do so. If possible, film holders, magazines, cassettes, etc., should be 
dried before use and should be loaded promptly and sealed with 
moisture-proof tape. In addition, they may be enclosed in one or 
more moisture-proof containers, capable of being sealed. Cameras 
should also be enclosed in moisture-proof containers, especially if 
they are loaded with film. Watertight slide fastener closures have 
recently been introduced which may prove to be well suited for use 
on moisture-tight bags. 

After exposure, the film holder should be temporarily sealed again 
until protective measures or final packing can be undertaken. 

Whenever possible, film handled under unfavorable conditions 
should be processed promptly, if necessary using a tropical processing 
procedure. Such processed film should then be adequately desic- 



June, 1946 USE OF DESICCANTS 501 

cated and sealed in a moisture-proof container to avoid growth of 
mold, fungus, or other damage. 

Special care should be given to motion picture film, once the manu- 
facturer's seal is broken, since lack of adequate equipment will often 
delay processing for many days or weeks under some field conditions. 
The raw film should be kept as cool as possible, although there is 
little practical benefit to be gained at temperatures below 50 F unless 
film is to be stored for many months. 

If the film has been exposed to high humidities, the protection of 
desiccation should be given unless it is to be developed immediately. 
In the event that the exposed film must be kept more than a week or 
two and that the storage temperature may exceed 85 to 90 F, it is 
advisable to dehydrate the film even if it has not been subjected to 
high humidity conditions. Any film which has been subjected to 
desiccating procedures should be rehumidified before handling to 
avoid markings caused by static electrical discharge. 

Practical Use of Refrigeration. Refrigeration equipment can be 
used effectively for desiccating film if conditions are set up so that 
the moisture given up by the film to the atmosphere in contact with 
it is removed therefrom as dew or frost on the cold surface. In order 
that the moisture content of the film shall be brought to the 40 per 
cent RH level which is recommended by one manufacturer the tem- 
perature of the cold surface should be 25 to 30 F lower than that of 
the film which is assumed to be in the range of temperatures above 
60 F. In order to ensure drying within about a day, the air should be 
in rapid circulation and should have access to all of the film surface. 
With rapid circulation it will be necessary that the air leaving the 
cold surface be reheated (to ambient temperature) so that it will not 
cool the film and thus lower the vapor pressure of the moisture in it. 
If the operation is carried on more slowly without circulation in an 
uninsulated container, the film probably will receive enough heat 
from the surroundings to maintain its temperature. The usual in- 
sulated domestic refrigerator, in which the cold surface is small and 
at a temperature below the frost point while the stored material 
more or less surrounds it and is at a temperature 15 or 20 F higher, 
can be used but will be slower. On the other hand, the film will have 
the protective effect of the lowered temperature during dehydration. 
There is no need for an insulated container for the drying operation. 
It is necessary only that an airtight (and probably light-tight) box 
enclose the film and cool surface and that provision be made for 



502 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

circulation and for keeping the film temperature high enough, prob- 
ably best above the dewpoint of the outdoor atmosphere to avoid 
having the film wetted as it is removed for repacking. There is no 
disadvantage in having the cold surface lower in temperature than 
that given above except that the precipitated moisture may be held 
as frost instead of draining away. 

Practical Use of Evacuation. Methods involving the use of a 
vacuum are of 2 types, those in which the pressure in the chamber 
is much less than the moisture vapor pressure corresponding to the 
desired condition of the film, and those in which it is higher, usually 
a few inches of mercury. In order to meet the 40 per cent RH re- 
quirement, the maximum allowable moisture vapor pressure of 
equilibrium at the end of the operation ranges from 0.56 in. of mercury 
for film at 90 F to 0.21 in. at 60 F. 

The procedures and equipment in the first-mentioned method are 
simple, consisting essentially of a pump, McLeod gage, and a tight 
evacuating chamber. This method is in many respects preferable to 
all other dehumidification methods but can be considered only when 
one-half to 2 hp is available either from a central service, a portable 
unit, or automotive take-off. No desiccants are needed in this 
procedure. It is worth some effort to have the use of this method if 
extreme conditions demand prompt action, especially with film un- 
favorably disposed as in wound rolls, film packs, etc. 

The specification of pumping equipment is somewhat involved 
because of the interdependence of working pressure, pump capacity, 
moisture load, and chamber size. In using this method it is suggested 
that the assistance of someone familiar with the practices of vacuum 
pumping be enlisted. In order to maintain the pressures of the order 
of one millimeter (0.040 in.) of mercury or a little lower necessary for 
a reasonably high rate of moisture extraction, the pump chosen is 
likely to be one offered by the laboratory supply houses for medium 
high-vacuum pumping. However, a large volumetric capacity in 
the range down to 0. 1 mm mercury is to be preferred over the ability 
of a pump to reach extremely low pressure as the pressure actually 
attained with a reasonable choice of equipment is likely not to range 
below 0.1 mm until most of the moisture is withdrawn from the film. 
Pumps which have a capacity of 10 to 20 liters per sec (20 to 40 cfm) 
should be capable of dehydrating a 1000-ft roll of motion picture 
negative film well below the safe upper limit in 5 to 8 hr. To avoid 
the uncertainties arising from estimates of moisture taken up pre- 



June, 1946 USE OF DESICCANTS 503 

viously by the film, it is advisable to pump long enough to remove 
most of the water which could possibly be present. The film will 
require rehumidification according to a procedure given later, before 
being handled again. 

Aspirator pumps and many of the ordinary vacuum pumps are 
incapable of maintaining a sufficiently low pressure for this 
purpose. 

Pressures of a Few Inches. In the second case where pres- 
sures of a few inches of mercury are employed, various types of 
mechanical vacuum pumps are acceptable. The reduction of pres- 
sure is used to remove most of the air so as to facilitate the movement 
of moisture through the space between the film and the desiccant 
which are in the evacuated enclosure. Since in this method the 
burden of absorbing the moisture given off by the film is taken entirely 
by the desiccant, a pump of small capacity is adequate and need not 
be kept in operation after the required pressure is attained. It might 
therefore even be of the hand-powered type. A single pump can serve 
a number of drying chambers if a valve is closed after evacuating 
and the pump transferred from one chamber to another. 

With the proper quantity of desiccant in this system, the moisture 
content can be reduced in 24 hr from any concentration to a safely 
low level even from motion picture film in a wound roll, as shown in 
Fig. 7 which was discussed in an earlier section. Unlike the pre- 
viously discussed lower pressure method, it is important in the present 
case that all obstacles to diffusion such as impervious wrappings be 
eliminated if retardation of the process is to be avoided. This 
method demands nothing beyond the small pump, the desiccant, and 
a working supply of tight containers, and should be adaptable to 
many situations. The quantity of desiccant required is the same as 
when normal pressure exists. 

Practical Use of Desiccants. It would seem that decisions as 
to desiccants and procedures may have to be made in the field 
under 2 conditions. In one, a photographic expedition might unex- 
pectedly find it necessary to keep exposed undeveloped film when 
operating under adverse atmospheric conditions. The need for de- 
hydration might be recognized but unless certain easily obtainable 
materials could be utilized, regardless of their efficiency, grave danger 
to the latent image might exist. 

In the second case, the need for dehydration might be anticipated 
when planning an expedition so that the choice of methods for com- 



504 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

bating unfavorable atmospheric conditions would be limited only by 
difficulty of transport or similar factors. 

In the practical use of desiccants, 2 steps are necessary: 

(1) An estimate must be made of the approximate weight of moisture to be 
removed. A knowledge of the relationship existing between equilibrium moisture 
content and relative humidity for the particular film being treated is useful. This 
relationship has been shown (Fig. 1) for a typical negative motion picture film. 

(2) The weight of desiccant required to reduce the moisture content to a de- 
sired value must be calculated (permanent packaging method) or the time neces- 
sary for an excess of a powerful desiccant to accomplish the same result must be 
estimated (temporary desiccator method). 

It is apparent that only the most approximate estimate of film 
moisture content is possible without elaborate instrumentation. 
However, in view of the fact that an error in the direction of overde- 
hydration has no severe penalty, provided the film is rehumidified 
before handling, such an estimate is entirely adequate. 

The factors which should be considered in estimating the quantity 
of moisture absorbed by the film have already been discussed in the 
sections following "Moisture Exchange and Equilibrium." Hence 
such materials as sheet film, small roll films, film packs, and similar 
products can be expected to approach moisture equilibrium with the 
relative humidity of the atmosphere to which they are exposed within 
a day under usual handling conditions. It is not unreasonable, 
therefore, to anticipate equilibrium with an atmosphere which is as 
high as 90 per cent RH with these materials. Motion picture film, 
on the other hand, in wound roll form has been shown to offer a dif- 
fusion path of high resistance even when the roll is wound loosely. 
It would therefore appear that a wound roll of undeveloped motion 
picture film given reasonable care in handling might attain equilib- 
rium with a relative humidity of only 80 per cent and usually less, 
in the same circumstances. In fact, if the factors of time, tempera- 
ture, and relative humidity were such as to cause the film to reach 
equilbrium with a relative humidity higher than 80 to 85 per cent, it 
is most likely that the film would simultaneously be damaged by mold. 

If conditions have been such that equilibrium with a relative 
humidity higher than 75 per cent is suspected, the more rapid act- 
ing, temporary packaging method or evacuation at low pressure is 
to be preferred. For an estimated equilibrium relative humidity of 
less than 75 per cent, the slower acting permanent packaging method 



June, 1946 USE OF DESICCANTS 505 

should prove adequate, unless temperature conditions are to con- 
tinue at 90 F or more. 

In view of the fact that a wound roll of motion picture film re- 
sponds more slowly than some other types of film to dehumidifying 
measures, the application of desiccating data already presented will 
be considered for that case. 

Temporary Packaging Method. On the assumption that a 
moisture content corresponding to a relative humidity of 85 per 
cent is the highest which is likely to be encountered with motion 
picture film, the minimum time required to reduce the moisture con- 
tent to relatively safe values can be determined. Fig. 7, Curve I, 
shows that under conditions similar to those which might be em- 
ployed in applying this method, using a large excess of a powerful 
desiccant, concentrated sulfuric acid in this case, approximately 2 
days would be required to reduce the moisture content from that of 
equilibrium with 85 per cent RH to equilibrium with about 40 per 
cent. The presence of wrappings would require more than if the film 
was not wrapped. 

Errors in overestimating the moisture content of the humidified 
film and thereby choosing a length of time for desiccating which is in 
excess of the minimum required for safe storage are not serious since 
the rate of moisture removal at levels corresponding to relative humid- 
ity values below 30 per cent is comparatively slow. This is shown 
in Fig. 7, Curve 1. 

Though detailed data relative to the rate of dehydration were not 
collected for other types of film in the course of the present work, 
similar considerations apply. These films should be arranged in a 
desiccator so that all surfaces are exposed to the drying action. The 
somewhat confined conditions existing in film packs or fnagazine 
wound miniature camera films would retard drying considerably. 
In such instances, parts of the film holder should, if possible, be re- 
moved to provide better access. 

The minimum weight of desiccant which can be used effectively 
with the temporary desiccator method depends principally on the 
moisture-absorbing capacity and rate of absorption of the desiccant 
being considered. While it is true that the use of a greater weight of a 
less active desiccant is capable of attaining the same rate of dehydra- 
tion as a smaller quantity of a more active agent, a practical limit is 
reached when the volume required is such as to create long diffusion 
paths. For instance, a deep bed of an agent, such as rice, would 



506 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

hardly produce an effect equivalent to that of a shallow bed of cal- 
cium chloride, even though a similar rate might be attainable if these 
agents could be utilized under equivalent conditions of access. 

As a practical guide to the use of desiccants with the temporary 
desiccator method, it appears that, with a large excess of an effec- 
tively utilized desiccant, the time required for the dehydration of 
mot on picture roll film is not less than 2 to 3 days; individual sheets 
of film may attain a satisfactory moisture content in 4 to 8 hr, while 
confined films (film packs, cassettes, etc.) should be treated at least 
one or 2 days, provided the moisture-proof cases confining such films 
are opened. 

Permanent Packaging Method. Probably the most direct 
way of illustrating the use of data already presented as it may be 
applied to the permanent packaging method is by the solution of a 
typical problem. 

Let us assume that it is desired to know the quantity of silica gel 
which will be required to reduce the moisture content of 1000 ft of 
negative motion picture film from that existing in equilibrium with 
80 per cent RH to that present at 40 per cent RH. 

On referring to Fig. 1 which shows the weight of water vapor ab- 
sorbed by 1000 ft of negative motion picture film in equilibrium with 
atmospheres of varying relative humidity, it is seen that the moisture 
content of the film at 80 per cent RH is about 83 grams; at the 
equilibrium condition desired, 40 per cent RH, it is 40 grams; there- 
fore the weight of moisture to be removed is 43 grams per 1000 ft of 
film. 

Referring now to Fig. 6, it is seen that at the desired terminal 
condition, 40 per cent RH, 0.175 gram of moisture will be absorbed 
by each* gram of silica gel. Since 43 grams are to be absorbed, the 
weight of silica gel required will be 43 divided by 0.175 or 246 grams 
(approximately l / 2 lb). 

This result for silica gel can be obtained more readily by reference 
to Fig. 8. By selecting the curve representing the initial moisture 
content of the film (specified in terms of the relative humidity with 
which it would be In balance), the quantity of silica gel required to 
reduce the moisture content to any other equilibrium condition can be 
read directly. 

Curves similar to those of Fig. 8 could be constructed for other 
desiccants by derivation from data presented in Figs. 1 and 6. It is 
interesting to note that if tea or rice had been used in the example just 



June, 1946 



USE OF DESICCANTS 



507 



given, approximately l l / 2 lb would have been required to attain the 
same equilibrium moisture content as was obtained with J / 2 lb of 
silica gel. However, since the rate of absorption with such agents is 
low, it would be advisable to increase the weight of desiccant used to 
assure an adequate rate of dehydration. An increased weight of 
desiccant would be used similarly to offset the retardation caused by 
the presence of a membrane, wrappings, etc. The fact that a lower 
equilibrium moisture content would eventually be obtained would 



2800 
2400 
2000 
1600 
1200 



600 



400 



70% 
60%- 



\ 




20 



30 40 



50 60 



70 



90 



TERMINAL PER CENT RELATIVE HUMIDITY 

FIG. 8. Amount of silica gel required to reduce mois- 
ture content of nitrate negative motion picture film in 
equilibrium with a given value of relative humidity to 
that of any other desired relative humidity. 

ordinarily be unimportant. The space requirement with tea, for 
example, would be increased also by its high specific volume. 

This practice can be extended to other films by determining the 
relationship existing between moisture content and equilibrium rela- 
tive humidity with the film being considered. 

Handling Desiccated Film. The effects of rewinding or other- 
wise handling unprocessed film which has been overdried are well 
known. One of the most serious of these effects is the markings 
caused by static electrical discharges. 

Because of the uncertainty involved in determining the actual 
moisture content of the film under field conditions, it is best to handle 
all film which has been subjected to dehumidification treatment as if 



508 C. J. KUNZ AND C. E. IVES Vol 46, No. 6 

it were overdried. Therefore, before such film is rewound or handled 
in any other way it should be rehumidified so that it attains equilib- 
rium with an atmosphere of 60 to 65 per cent RH. One method for 
accomplishing this is as follows : The roll or the sheets of film should 
be removed from the sealed container in the darkroom with care to 
prevent cinching or sliding. Film in film packs or in cassettes should 
not be removed from the holder until after it is humidified, although 
parts of the holder not in contact with the film which might impede 
diffusion should be removed. The film should then be laid on an open 
grill or screen arranged so that air can circulate both above and below. 
The humidifying air should be at 60 to 65 per cent RH (not higher) 
and should be circulated in the vicinity of the film by means of a 
fan. 

The time required to rehumidify a wound roll of motion picture 
film depends, of course, on its initial moisture content. However, 
under the conditions described, one day (24 hr) should be sufficient to 
rehumidify such film, even though it has attained equilibrium with a 
low value of relative humidity. Failure to circulate the air increases 
the time required many times. Other types of film would normally 
respond somewhat more quickly than wound motion picture film be- 
cause of better access to the humidifying atmosphere. 

Care of Desiccants. It is necessary that the desiccant be stored 
in a sealed container up to the time of use. Products sold specifi- 
cally for the purpose are ordinarily fully activated and are supplied 
in suitable containers by the manufacturer. Obviously, provisions 
must be made for resealing a large stock container if only part of its 
contents are removed at one time. A better plan is to provide many 
small packages so that the unused portion of the stock will remain 
unaffected. 

If it becomes necessary to reactivate used desiccant, or to activate 
such substances as tea or rice, this can be accomplished in an oven 
such as is used to bake or roast foods (bread or meat) . Ovens designed 
for reactivating are generally intended for large-scale operations and 
require a large source of heat or electrical power. Inorganic desic- 
cating agents should ordinarily be heated to a temperature between 
300 and 500 F. One or 2 hr treatment at the higher temperature and 
4 to 5 hr at the lower temperature should be sufficient. Inorganic 
desiccants which have become contaminated with oil or other organic 
matter should not be reactivated in the field because there is no pro- 
tection against fire when the material is heated. 



June, 1946 USE OF DESICCANTS 509 

Some organic agents, such as tea or rice, however, cannot be heated 
much above 250 F without decomposing; this is indicated by the 
evolution of an abundance of smoky fumes. Rice may be considered 
ready for use when, on stirring during heating, it becomes uniformly 
light brown in color. Photographic black paper has been heated 
nearly to 500 F before smoking, while one grade of white paper 
scorched within one to 2 min at about 600 F. 

Certain commercial desiccants are available in an "indicating" 
form that is, they are impregnated or coated with an agent which 
changes color as the moisture content exceeds a critical value. Co- 
balt chloride is sometimes used as an indicator, undergoing a color 
change from blue to pink as moisture is absorbed. The transition oc- 
curs in the range of moisture content corresponding to 30 to 60 
per cent RH. Therefore the mere fact that the color is blue does 
not necessarily indicate that the desiccating agent is fully activated; 
however, a neutral or pink color does indicate that, for practical pur- 
poses, the drier is entirely spent. 

Certain transparent or window-type packages supplied commer- 
cially are not intended to withstand the high temperatures used in 
reactivation. 

Improvised methods of activating, such as drying the intended 
desiccant on a shovel held over a fire, or pouring it over heated rocks, 
have sometimes been described. Such methods are capable of limited 
success but results are rather uncertain. It would seem that they 
should be applied only if the need for desiccating had not been fore- 
seen when planning an expedition and the use of available materials 
and methods was later found necessary. The following suggestions 
may be found useful in this situation. 

Activation should be carried on so that the temperature of the 
mass of desiccant is above the boiling point of water. While glass 
thermometers can be obtained for temperatures up to 400 F, and 
all-metal-type thermometers up to 1000 F, it seems likely that, under 
unforeseen conditions, such equipment would be lacking. In this 
case it can be made certain that the temperature of boiling water has 
at least been attained by embedding a few small pebbles in the desic- 
cant; when these are removed and dropped into water, momentary 
sizzling indicates that a temperature of at least 212 F has been 
reached. In some cases, lead-tin solders may be available. These 
have their melting point in the range of temperatures between 212 
and 500 F and therefore may be used for indicating temperatures. 



510 C. J. KUNZ AND C. E. IVES 

The temperature of activation should not exceed that at which de- 
composition of the desiccant takes place. 

The temperature within a mass of the desiccant during treatment 
rises slowly because the material is a poor heat conductor and heat is 
taken up in the process of evaporating water. For these reasons the 
oven air temperature can be considerably higher than the safe upper 
limit for the dessicant. If the temperature of the bed increases 
rapidly later in the drying process, it is a likely indication that evapo- 
ration has ceased and that the agent is substantially dried. 

During treatment, the material should be spread out in a layer 
1 to 2 in. thick, if possible. Two to 4 hr treatment should be suf- 
ficient under suitable temperature conditions. After the heating 
has been completed, the substance should be transferred immedi- 
ately to a sealed container and allowed to cool there before use. 



AMERICAN STANDARD 16-MM TEST FILMS 

As part of its wartime standardization work, the Society of Mo- 
tion Picture Engineers prepared rigid specifications for nine 16-mm 
test films to meet fundamental test and performance requirements 
laid down by our armed forces. 

Excessive equipment failures and the lack of suitable test standards 
for 16-mm projection were the reasons behind the request of the 
services to both the Society and the Research Council of the Academy 
of Motion Picture Arts and Sciences for aid in this War Standards 
project. During the war and shortly thereafter, most of the test 
films produced were required for government obligations, but now 
that the service contracts have been substantially completed, it is 
possible for the Society to begin supplying these films to civilian 
industry in this country and abroad. 

Six of the American War Standards, 5 specifying test films and 
one covering a projection type of resolution test plate, were reviewed 
under the regular procedure of the American Standards Association's 
Sectional Committee on Motion Pictures Z22, in October, 1945. 
They were approved by Z22 and subsequently by the SMPE to con- 
tinue as American Standards rather than as American War Stand- 
ards. The Z52 number designations have been superseded by Z22 
numbers and all 6 standards appear in the April, 1946, JOURNAL in 
American Standard format. The 6 standards specify 6 individual 
test films and the one resolution test plate, as listed below. 

For the convenience of those who worked on, or are otherwise 
familiar with, the War Standards, the Z52 numbers are shown in 
the list that follows as subordinate references. They should not, 
however, be used when ordering either the standards or the films. 

The printed standards are now available from the general offices 
of the Society, either as individual sheets at 15 cents each, or in 
combination with the other 14 American Standards which were pub- 
lished in the April, 1946, JOURNAL. The latter are furnished in an 
attractive leatherette covered, loose-leaf binder at $4.50 per set. 
The binder may be purchased separately at $2.50; however, the 
combination of binder with all of the current American Standards 

511 



512 AMERICAN STANDARD TEST FILMS Vol 46, No. 6 

represents a saving of $1.00 per set. (All subsequent American 
Standards on Motion Pictures will be issued punched to fit this binder 
and may be purchased either from the American Standards Associa- 
tion, 70 East 45th Street, New York, N. Y., or the Society of Mo- 
tion Picture Engineers, Hotel Pennsylvania, New York 1, N. Y.) 

The American Standard Test Films are made to extremely close 
tolerances, certain of them are individually calibrated, and all are 
packed in sealed metal containers with complete instructions for use. 
They are available from the general offices of the Society, and the 
prices quoted include shipping charges within the United States. 

Sound Focusing Test Film. Sound Focusing Test Films for 16-Mm Sound 
Motion Picture Projection Equipment, Z22. 42-1946 (Z52.ll*). Service Type, 
100 ft at $27.50; Laboratory Type, 100 ft at $27.50. 

This test film carries a special "square wave" track, chosen because its output 
changes more rapidly with changes in the focus of the sound optical system of the 
projector than the output from the usual "sine wave" high-frequency track. The 
"square wave" track also gives a more sensitive indication of the errors of the 
"azimuth" adjustment of the sound reproducing light beam. 

The Sound Focusing Test Film is an original negative and is made in 2 types: 
Laboratory Type, being a 7000-cycle record for manufacturing and precision ad- 
justment of the focus and azimuth of the sound optical system, and Service Type, 
being a 5000-cycle record for quick service adjustment. 

3000-Cycle Flutter Test Film. 3000- Cycle Flutter Test Film for 16-Mm Sound 
Motion Picture Projection Equipment, Z22. 43-1946 (Z52.9 *) . 380 ft at $104.50. 

This test film is a direct-positive original recording that carries a 3000-cycle 
record having extremely low nutter content for use in measuring the nutter intro- 
duced by 16-mm sound reproducers. The recorded frequency is within 25 
cycles of the 3000-cycle frequency, the output level is constant within x /4 db, 
and the total flutter content of the film at the time of shipment is less than 0.1 per 
cent. 

Multifrequency Test Film. Multifrequency Test Film for Field Testing 16-Mm 
Sound Motion Picture Projection Equipment Z 22. 44-1946 (Z52.8*). 150 ft at 
$41.25. 

This test film contains the following series of frequencies each preceded by a 

spoken announcement: 

Frequency Cycles 

400 2000 

50 3000 

100 4000 

200 5000 

300 6000 

500 7000 

1000 400 

* The Z52 number is included for reference only and is not a part of the Ameri- 
can Standard title. 



June, 1946 AMERICAN STANDARD TEST FlLMS 513 

This is a direct-positive original recording and the several frequencies are re- 
corded in such a manner that if reproduced by a scanning light beam of negligible 
width, they would produce constant light modulation at the phototube. 

Each film is individually calibrated on equipment correct within =*=V4 do up 
through 3000 cycles and within =*= l / 2 db above 3000 and through 7000 cycles. 
The deviation from the intended flat response characteristic (assuming negligible 
reproducing light beam width) is stated as a correction for each frequency which 
will give the true level when it is added algebraically to the output level measure- 
ment obtained when using the film. 

400-Cycle Signal Level Test Film. 400- Cycle Signal Level Test Film for 
16- Mm Sound Motion Picture Projection Equipment, Z22. 45-1946 (Z52.17*). 
100 ft at $27.50. 

This film is a direct-positive original recording designed to furnish as nearly as 
is practical an absolute standard of recorded signal level for use in measuring 
the effective amplification and sound output of 16-mm sound motion picture pro- 
jectors taking into account the sound optical system and phototube, as well as the 
amplifier and loudspeaker. 

A definite output level is determined by specifying the amplitude of the re- 
corded signal, the density of the image, and the combined base and fog density 
of the clear part of the sound track within narrow limits. The specified level is 
approximately 2 db below the maximum level possible and is about equal to the 
highest level that is to be expected in most recording, since in commercial practice 
the image density is usually not so great and the fog density not so low as the 
values specified for this film. 

The actual measured values of signal amplitude, image density, and fog density 
are given with each film, together with the corresponding calculated value of over- 
all deviation from the intended standard signal level. 

Resolving Power Test Target.- Method of Determining Resolving Power of 
16-Mm Projector Lenses, Z22. 53-1946 (Z52.5*) . Glass test plate at $10. 

This target is a high resolution glass plate carrying a series of 19 groups of line 
patterns suitably spaced within the standard 16-mm projection aperture area. 
Each pattern consists of 7 sets of parallel lines indicating lens resolving power of 
from 20 to 90 lines per mm. This plate is intended for use with a special test 
projector designed to hold both the test target and the 16-mm projection lens 
under test. The resolution of the lens may then be read directly from the pro- 
jected image on a screen. 

Travel Ghost Test Film. Method of Determining Freedom from Travel Ghost in 
16-Mm Sound Motion Picture Projectors, Z 22. 54-1946 (Z52.4 *) . 100 ft at $12. 

This film carries a pattern consisting of 9 small diamond-shaped transparent 
areas on a black background. Three of these areas are at the top of the picture 
near the frame line, three of them are on the horizontal center line of the frame, 
and three are near the frame line at the bottom. 

* The Z52 number is included for reference only and is not a part of the Ameri- 
can Standard title. 



514 AMERICAN STANDARD TEST FILMS Vol 46, No. 6 

Light streaks (travel ghost) running upward or downward on the screen from 
the projected diamond-shaped areas show that the projector shutter is improperly 
adjusted. With a properly designed and properly adjusted 16-mm projector, no 
travel ghost will be seen. 

AMERICAN WAR STANDARD 16-MM TEST FILMS 

There are 5 American War Standard Test Films now available 
to the industry in this country and abroad from the Bell and Ho well 
Company, 7100 McCormick Road, Chicago, 111.; the Research 
Council of the Academy of Motion Picture Arts and Sciences, 1217 
Taft Building, Hollywood, Calif.; and the Society of Motion Pic- 
ture Engineers, Hotel Pennsylvania, New York 1, N. Y. These 
5 test films are specified in detail in the 4 American War Standards 
listed below. Printed copies of these War Standards are available 
only from the American Standards Association, 70 East 45th Street, 
New York, N. Y., at 10 cents each. 

Selected Sample Test Film. Test Film for Checking Adjustment of 16-Mm 
Sound Motion Picture Projection Equipment, Z52.2-1944. 

This is a sound test film with accompanying picture for checking the adjustment 
of 16-mm sound motion picture projectors and for judging the acoustics of the 
room in which the projectors are operated. The film is a 16-mm print rere- 
corded from a 35-mm original, containing main title music, 3 excerpts from studio 
feature releases recorded on variable-density track, and 3 excerpts from studio 
feature releases recorded on variable-area track. There is also a piano section 
recorded on variable-area track. 

This film is 380 ft in length, and is available from the Research Council of the 
Academy of Motion Picture Arts and Sciences, 1217 Taft Building, Hollywood, 
Calif., at a price of $30 each. However, an inquiry concerning delivery 
should be directed to the Research Council before placing an order. 

Picture Unsteadiness Test Film. Method of Determining Picture Unsteadiness 
of 16-Mm Sound Motion Picture Projectors, Z52.6-1944. 

This is an opaque film with circular punched holes of uniform diameter pre- 
cisely located in the picture area. Picture jump and weave is determined di- 
rectly by projecting the image of these holes against vertical and horizontal 
scales placed at the projection screen. 

This test film is 100 ft in length and may be purchased from the Bell and Howell 
Company, 7100 McCormick Road, Chicago, 111., at $11 each. 

Scanning Beam Illumination Test Film. Method of Determining Uniformity of 
Scanning Beam Illumination for 16-Mm Sound Motion Picture Projectors, Z52.7- 
1944. 

This test film carries a narrow sound track (0.005 in. wide) modulated at con- 
stant level by a 1000-cycle tone. The location of this sound track changes at a 
uniform rate along the length of the film from a position just inside one edge 
of the scanned area to a position just inside the opposite edge of the scanned area. 



June, 1946 



AMERICAN STANDARD TEST FILMS 



515 



The narrow 1000-cycle sound track sweeps across the scanning light beam from 
one end to the other at a uniform rate, the position of the sound track relative to 
the ends of the light beam at any instant being shown by an animated diagram 
appearing in the picture area. 

If the scanning beam illumination were absolutely uniform across the width 
of the scanned area, the output level of the 1000-cycle tone would be constant. 
In practice, however, some variation of an output meter reading will always be 
observed. By running a loop of the film continuously and observing the indica- 
tions of the output meter while adjustments are made, it is usually possible to cor- 
rect unevenness of illumination and bring the variation of output within a limit of 



1000 CYCLE5- 



-300 CYCLES 



FIG. 1. Buzz track. 



The Service Type may be spliced into continuous 3Y 2 -ft loops and the Labora- 
tory Type may be spliced into 34-ft loops. Each type is 100 ft long and is priced 
at $27.50. They are available from the Society of Motion Picture Engineers. 

Buzz Track Test Film. Buzz Track Test Film for 16- Mm Sound Motion 
Picture Projectors, Z52. 10-1944. 

This test film is an original negative that carries a track as shown in Fig. 1. 
It is made on a special recording machine, is extremely accurate in position, and 
almost entirely free from weave. 

When the film is run on a projector in correct adjustment and free from weave, 
no sound is heard. Either or both the 1000- and 300-cycle tones will be heard; 
however, if the scanning light beam is out of adjustment. 

This film is available from the Society of Motion Picture Engineers in 100-ft 
lengths at $27.50. 



STANDARDIZATION AND THE ANTITRUST LAWS* 
JAMES D. HAYES** 



Summary. The following discussion and interpretation of recent changes in 
the attitude of the courts and the Federal Trade Commission toward standardization 
was presented before the first meeting of the Conference Committee of Staff Executives 
of Member Bodies and Associate Members of the American Standards Association 
on March 7, 1946. The author is a recognized authority on the legal aspect of indus- 
trial standardization, and the editors of the JOURNAL feel that this frank discussion 
will be of particular interest to members of the Society who are manufacturers, or 
who have participated in the 30 years of standardization history of the SMPE. 

As a result of the first World War and the tremendous demands 
made upon industry, standardization became an increasingly im- 
portant cooperative activity of industry, and the Department of 
Commerce lent all its influence to promoting industry activity in this 
field. In 1922, Herbert Hoover, as Secretary of Commerce, re- 
quested an informal opinion of the Attorney General as to the legality 
of trade association activity in general and included in his request the 
following question : 

"May a trade association, in cooperation with its members, advocate and 
provide for the standardization of quality and grades of product of such mem- 
bers, to the end that the buying public may know what it is to receive when 
a particular grade or quality is specified; and may such association, after 
standardizing quality and grade, provide standard form of contract for the 
purpose of correctly designating the standards of quality and grades of product; 
and may it standardize technical and scientific terms, its processes in produc- 
tion, and its machinery; and may the association cooperate with its members 
in determining means for the elimination of wasteful processes in production 
and distribution and for the raising of ethical standards in trade for the pre- 
vention of dishonest practices?" 

This was the first occasion upon which any authority had ventured 
an opinion on the legality of standardization, or rather, it was the first 
occasion upon which legality, theretofore assumed, had been ques- 

* Reprinted from Industrial Standardization, 17, 4 (Apr., 1946), p. 74. 
** Member of the firm, Donovan, Leisure, Newton, Lumbard, and Irvine. 
516 



STANDARDIZATION AND ANTITRUST LAWS 517 

tioned. The Attorney General did not give a very satisfactory 
answer to Mr. Hoover's inquiry in the light of subsequent develop- 
ments. He said : 

"I can now see nothing illegal in the exercise of the other activities men- 
tioned, provided always that whatever is done is not used as a scheme or device 
to curtail production or enhance prices, and does not have the effect of sup- 
pressing competition." 

It is obvious that with such hedging almost any group activity would 
be proper under the antitrust laws. 

Accepted as Green Light for Standardization. Be that as it may, 
this correspondence between the heads of the Department of Com- 
merce and the Department of Justice was accepted as giving a green 
light to technical standardization activity by industry groups and, 
therefore, for several years, it was often referred to by the Department 
of Commerce in its efforts to encourage industry activity in this field. 

The first, and almost the only, mention made by the Supreme 
Court of standardization activity occurred in 1925 in the familiar case 
of Maple Flooring Manufacturers Association versus United States. 1 
In that case a price-fixing conspiracy was charged against maple 
flooring manufacturers. One of the means alleged to have been used 
to effectuate the conspiracy was standardization of grades of flooring. 
A decree against the Association was entered by the district court, 
but the decree made no reference to the standardization activities of 
the Association. On appeal to the Supreme Court, the Court in 
reversing the decree of the lower court took notice of the Association's 
standardization activities. The Court said: 

"The defendants have engaged in many activities to which no exception 
was taken by the government and which are admittedly beneficial to the 
industry and to commerce; such as cooperative advertising and standardiza- 
tion and improvement of the product/' 

In its discussion of the facts it made no other reference to standardiza- 
tion. At the least, the Court thought standardization beneficial to 
industry and commerce. However, it is important to note in this 
case, as I shall point out more fully, that the Association was found 
not guilty of any price-fixing conspiracy. 

In the years following this decision, standardization by joint action 
of industry thrived, and as far as can be determined no question was 
raised as to its legality. 

Standardization Legal; Illegal When Used for Illegal Purposes. 
In 1923, a consent decree was entered in a case in which the govern- 



518 J. D. HAYES Vol 46, No. 6 

ment complained against the Tile Manufacturers' Credit Association, 
et a/., 2 alleging a price-fixing agreement. The complaint listed 17 
practices as the subject of the conspiracy. Among these was the 
allegation that the defendants conspired "to standardize the shapes, 
etc., of tile made, eliminating many now sold, and establish the use of 
standardized catalogs of said association (catalogs now being pre- 
pared)." 

The decree provided that nothing therein should be deemed to re- 
strain the defendants from maintaining an association for certain 
enumerated purposes including "to secure and maintain the standard- 
ization of quality and of technical and scientific terms, and the 
elimination of nonessential types, sizes, styles or grades of products." 

On the other hand, the consent decree entered versus the Carpet 
Manufacturers of America, Inc., in New York in 1941 prohibited 
agreements "to limit the kinds, quality, grade, quantity or the 
number of lines of merchandise to be manufactured and sold." 

The conclusion to be drawn from these cases was that standardiza- 
tion when used as a means to further, a price-fixing conspiracy or a 
restraint of production or an elimination of competition was illegal 
whereas standardization, as such, was unobjectionable. This con- 
clusion was buttressed by various pronouncements of the Federal 
Trade Commission, the chief one being a statement by the chairman 
of the Commission, made in 1931 that "in no matter has the Commis- 
sion ever held standardization of commodities by members of an 
industry to be violative of any of the statutes it has the duty of en- 
forcing." 

Perhaps the FTC has had a change of heart. Since 1938, hardly a 
complaint involving trade associations has been issued by the Com- 
mission which did not allege standardization as one of the means 
utilized in advancing and perfecting the alleged conspiracy. 

The convenient division of the authorities theretofore thought pos- 
sible was, however, made somewhat questionable by the Milk Can 
Institute case. The original complaint in that case was issued in 
June, 1934, under the title of Keiner Williams Stamping Co., et al. z 
It contained an allegation that in furtherance of an alleged competi- 
tion-suppressing, and price-fixing conspiracy the respondents (mem- 
bers of a trade association) had standardized the construction of milk 
and ice cream cans so that they were of uniform material, weight, 
and general construction. Following this allegation appeared a 
caveat to the effect that "The Commission is not here complaining 



June, 1946 STANDARDIZATION AND ANTITRUST LAWS 519 

against the alleged standardization as such, but only against the use 
thereof as a means of carrying^ut the price-fixing conspiracy herein- 
before charged." 

No decree was ever entered on this complaint and in July, 1941, it 
was dismissed without prejudice and a new complaint was issued. 
The matter was now entitled "In the Matter of the Milk and Ice Cream 
Can Institute, et a/." 4 The amended complaint also alleged a price- 
fixing conspiracy and as one of the activities engaged in "pursuant to 
and in furtherance of the aforesaid combination," the elimination of 
models and styles of cans, the change in designs of cans and other 
standardization of products independently of and beyond any require- 
ments for standardization prescribed by the Federal or State Govern- 
ments "for the purpose of eliminating competition in the attractiveness 
of their products to buyers " Significantly the caveat contained in the 
original complaint was missing from the amended complaint. 

FTC Attacks Standardization as Such. It is particularly to be 
noted that the language quoted indicates that for the first time the 
Federal Trade Commission was attacking standardization as such. 
Any standardization program naturally results in some elimination of 
competition of attractiveness of product and in some restraint of 
production in that it standardizes the product available to prospective 
purchasers. But that end has always been thought to be a justifica- 
tion for standardization. In other words the benefit of standardiza- 
tion to the consumer arises from the very fact that the product is 
standardized, enabling him to buy with confidence and giving to him 
the advantage of low prices owing to savings in manufacturing costs 
and interchangeability of parts. These advantages have long been 
thought to far outweigh any incidental restraint of production arising 
from the elimination of special items. 

Following hearings in the Milk Can case, the Commission made 
findings of fact and conclusions of law. Among these was a finding 
that the respondents had engaged in standardization and simplifica- 
tion "as a further means of establishing a basis upon which price 
differences might be eliminated, and for the purpose of eliminating 
competition in the attractiveness of their products to buyers." The 
findings set forth an example of the standardization activity of the 
Institute which, so far as appears, is a typical association activity. 
The specific language of the findings is : 

"At a meeting held June 14, 1932, the respondent, D. S. Hunter, as com- 
missioner, called attention to the desirability, in the work of standardization, 



520 J. D. HAYES Vol 46, No. 6 

of eliminating if possible some styles and sizes of milk and ice cream cans, 
especially those for which there was a small demand, and also that considera- 
tion should be given to standardizing theSveight, as well as the gages, of the 
various styles and sizes of cans. The Commissioner was instructed to com- 
municate with members to determine what lines of cans could be eliminated. 
Subsequent thereto, at a meeting held on July 12, 1932, the committee on 
standards submitted a table of standardization of various styles and sizes of 
cans by weight and on motion made, seconded, and carried, this recommenda- 
tion by the committee on standards was adopted by the respondent members. 
Subsequent to that time, the committee on standards had made various 
recommendations with reference to gages and weights of milk and ice cream 
cans which were adopted by the respondent members." 

The order to cease and desist entered by the Commission pursuant 
to the hearings and findings made no specific reference to the stand- 
ardization activity of the Institute. It did specifically prohibit the 
members of the Institute from fixing or maintaining prices or adhering 
to such prices and it prohibited a sales and price-reporting service 
and a freight equalization system used in connection with price fixing, 
etc. The decree did contain a general paragraph prohibiting the 
members from engaging in "any other practice or plan which has the 
purpose or effect of fixing or maintaining prices. . . " 

The Institute and its members appealed from the order of the Com- 
mission to the Seventh Circuit Court of Appeals and on January 7, 
1946, the Circuit Court rendered its opinion upholding with minor 
modifications the Commission's order. 

The court treated the standardization activity of the Institute and 
its members as evidence of the conspiracy. The court noted that the 
only basic question before it was whether the members had conspired 
to fix prices. The court discussed the standardization activities of 
the defendants as being one of the pieces of evidence supporting the 
Commission's finding that the defendants had acted in concert and by 
agreement. 

There was no discussion or intimation by the court to show that it 
thought that standardization by itself would result in a restraint of 
trade, but its whole discussion was directed to the point that the mem- 
bers had agreed on standardization and that this agreement sup- 
ported the Commission's finding of a conspiracy. The court said : 

"We think it is true that they were standardized in the instant situation, 
but this was the result of the activities of the Institute and its members. In 
fact, there was a continuing effort and urging on their part that the cans be 
manufactured in uniform classifications. It may be, as argued, that much 
of this effort was to comply with various governmental regulations and for 



June, 1946 STANDARDIZATION AND ANTITRUST LAWS 521 

health purposes, but the fact still remains that it was easier to reach the goal 
of uniform prices on a standard product than on one which was not. The 
meticulous effort disclosed by the record by which petitioners standardized 
their products is also a strong circumstance in support of the Commission's 
findings that their activities were the result of an agreement." 

It is true that in the Milk Can case there were the usual other fac- 
tors present which theretofore had been present in every case where 
the Commission had attacked standardization, namely, a price-fixing 
conspiracy, restraint of production, and elimination of competition. 

The language of the Federal Trade Commission findings was that a 
table of standardization "was adopted" by the members of the Insti- 
tute. There is nothing either in the findings, the order to cease and 
desist, or the decree of the Circuit Court of Appeals to indicate what 
the members did when they "adopted" the table of standardization 
recommended by the Milk Can Institute Committee on Standards. 
It would seem fair to conclude that what was complained of was 
merely the fact that the members agreed to adopt certain sizes, styles, 
and .types of cans as standard and that there was no agreement to 
adhere to the standardized line. If there had been both, the Com- 
mission and the Circuit Court of Appeals had many authorities on 
hand to support a holding that such activity was a restraint of trade. 
Since no reference was made to an agreed restraint it seems fair to 
assume that there was none. 

On the foregoing assumptions one must inevitably conclude that 
the court utilized an agreement theretofore considered perfectly 
legitimate, namely, an agreement on what the standard shall be, to 
convict the defendants of a price-fixing conspiracy. Therefore, under 
this case, no longer will legality of standardization depend upon 
whether the program is misused and made a part of restraint of trade, 
but on whether the Association happens to be prosecuted for a re- 
straint of trade arising from any one or more of its other activities. 

Prior to the Milk Can case, most lawyers active in the trade associa- 
tion field felt safe in advising trade associations that activity in the 
field of standardization was legitimate as long as the standards were 
arrived at on sound engineering and technical considerations (as well 
as considerations of safety and public health) as long as the activity 
was carried on in good faith and without any intent to fix prices, 
restrict production, and eliminate competition, and finally, as long as 
there was no agreement express or implied among the members to 
adhere to the standards agreed upon. 



522 J. D. HAYES Vol 46, No. 6 

It is to be noted that the Attorney General, back in 1922, based his 
clearance of standardization on the premise that the standards were 
voluntary and that any member was left free to adhere to or depart 
from standards as he might from time to time see fit. It was never 
thought to be illegal to agree on what the standards should be. In 
fact it is obvious that no standardization can be carried on except on 
the basis of agreement as to what the standards shall be. 

Find Danger in Agreement on What Standard Shall Be. Now, 
however, it appears that there is danger even in an agreement on what 
the standard shall be in that the Federal Trade Commission and the 
courts will accept that agreement as evidence of agreement on other 
matters upon which the law forbids competitors to agree. 

I mentioned above that most Trade Commission cases since 1941 
against trade associations have contained an allegation that standardi- 
zation was one of the means utilized to promote the conspiracy. It is 
interesting to note in the Milk Can case that the Commission found 
that the standardization work was done "as a further means of 
establishing a basis upon which price differences might be eliminated 
and for the purpose of eliminating competition in the attractiveness 
of their products to buyers." This limitation does not mean much 
when one remembers that the Commission can always make such an 
allegation after the fact. In other words, if a restraint of trade exists, 
a standardization program can always be pointed to as a means of 
effectuating the restraint. There never has been any doubt that 
price fixing or restraint of production is facilitated by uniformity of 
product. 

The only comforting thing about the Milk Can decision in the 
Seventh Circuit Court of Appeals is that the court deleted the general 
catch-all provision of the order to cease and desist, thereby leaving the 
order without any possible prohibition against the standardization 
activity of the Institute. 

Summary of Present Legal Status of Standardization. To sum- 
marize the present state of the antitrust law as it applies to standardi- 
zation : (7) the activity is legitimate in the absence of any agreement 
by the participants to limit their production to the standard items; 
(2) if an association or other group is prosecuted or complained 
against for a conspiracy to restrain trade either by fixing prices, 
restraining production, or eliminating competition, or any of the vari- 
ables or combinations thereof, it is almost certain that their stand- 
ardization activity will be cited as evidence of the conspiracy ; ( 3} it 



June, 1946 STANDARDIZATION AND ANTITRUST LAWS 523 

is highly improbable that a trade association or other group will ever 
be prosecuted or complained against solely for carrying on a standardi- 
zation program in the absence of any other restraint of trade. 

Since the attack on standardization in the Milk Can case was ac- 
companied by allegations of price fixing and restraint of production 
by means of other more serious methods, one cannot state positively 
what the result would be of a complaint addressed to standardization 
alone. I do not believe that a court would hold that trade and com- 
merce had been restrained by the incidental elimination of "competi- 
tion in the attractiveness of products to buyers." I think the courts 
would hold that such restraint, if any, was a restraint of trade under 
the rule of reason. 

Whether the Federal Trade Commission will ever bring a com- 
plaint based upon that type of restraint alone and what conclusion the 
Commission would reach remains to be seen. The language in the 
Milk Can case indicates that the Commission believes standardization 
for the purpose of eliminating competition in the attractiveness of 
their products to be a violation of law. It is impossible to tell from 
the Milk Can case whether the Commission really believes that. One 
word of warning is appropriate. If the Commission should bring a 
proceeding and issue a cease and desist order based upon such a re- 
straint, it is doubtful whether the court would upset an order of the 
Commission in view of the well-known rule that the Commission's 
conclusions are conclusive on appeal if supported by substantial 
evidence. 

"Standardization" Needs Exact Definition. One of the greatest 
sources of confusion and doubt on this question of the legality of 
standardization arises from the looseness with which the term 
"standardization" has been used. Perhaps I should have defined 
the term at the beginning of my talk, but it seemed to me that it 
would be more advisable to leave you with a proper, exact definition. 
A standard is a "definition of a product or process with reference to 
composition, construction, dimension, quality, operating characteris- 
tics, performance, nomenclature, and other like factors." And 
standardization, as I have used it here, is "the formulation of such 
definition or standard." Standardization has nothing to do with an 
agreement to adhere to the standard so formulated and it behooves 
every association engaged in this type of activity to make clear be- 
yond question both to its members and to the public generally that 
the promulgation of a standard or standards by it does not preclude 



524 J. D. HAYES Vol 46, No. 6 

any member or nonmember from making his own determination as to 
whether or not he will manufacture in accordance with the standard. 

Of course, it is natural that a manufacturer who has taken part in 
setting up standards will in all probability manufacture in accordance 
with the standards. As long as each manufacturer freely, voluntarily, 
and in good faith does this, whatever restraint of trade results is 
incidental and is far outweighed by the economic benefits accruing to 
consumers, distributors, and manufacturers from a standardization 
program. It is only where the freedom of the individual has been 
taken from him by an agreement express or implied to adhere that 
the restraint becomes direct rather than incidental. 

Voluntary Nature of Standardization Eliminates Use of Com- 
pulsion. A natural corollary of the voluntary character of stand- 
ardization is that an association engaged in standardization activity 
take no steps to compel compliance with its standards. This, of 
course, does not mean that the association cannot investigate to see 
whether its standards are being followed for the purpose of ascer- 
taining and verifying the validity of the standards and the possible 
necessity of amendment. Such activity is proper if carried on in good 
faith. Members are not likely to be confused as to the real purpose of 
any investigation along these lines. 

How the Party-at-Interest Procedure of ASA Affects Legality. 
There is a question related to the voluntary use of standards of special 
interest to this group. This question has not yet been answered by 
legal decision, but in my opinion it can have but one answer. That 
question is, "Does the procedure of the American Standards Associa- 
tion, in which all parties at interest are represented in the develop- 
ment of a standard, change the picture as far as the legality of a 
standard is concerned?" Although there has been no legal ruling on 
such a question, it is my opinion that this procedure does give a 
standard greater protection than if the standard is developed by a 
private group. If a standard is arrived at after consulting not only 
the interests of the manufacturers but also of the marketers of the 
product, of the consumers of the product, and of the government 
agencies, if any, which may be interested in the use or disposition of 
that product, then whatever restraint results from the adoption of 
that standard would, in my opinion, be reasonable. I do not believe 
it could ever be successfully attacked. 

One should not infer, however, from the fact that a government 
agency has participated in any way in the adoption of a standard that 



June, 1946 STANDARDIZATION AND ANTITRUST LAWS 525 

the standard so adopted carries with it any license to violate the law in 
the use of that standard. This is true even though the government 
agency might request that the particular wrongful conduct be done. 
If, for example, a government agency should ask an industry to adhere 
to the standard and not to manufacture articles which do not comply 
with the standard and if, pursuant to such request the industry should 
so agree, the agreement thus made would still be subject to attack. 
The reason is that no government official no matter how highly placed 
has the authority to authorize any individual or group of individuals 
to violate the law or to invite them to do so. 

Confusion Equally Great on Status of Simplification. Finally, I 
should say a word about simplification. The confusion on this sub- 
ject is even more extensive than on standardization and again arises 
primarily from the wide variety of activities that has been called 
simplification. Everything that I have said regarding standardiza- 
tion applies to simplification, as I define simplification. 

Simplification is "the formulation of standard product lines con- 
sisting of types, sizes, shapes, grades, colors, and varieties of product 
most frequently demanded by consumers." 

Same Limitations Apply to Simplification. What I have just 
defined as simplification is sometimes called type standardization. 
It has also been defined as an agreement to eliminate in accordance 
with agreed product lines. The same limitations that apply to 
standardization also apply to simplification. It is legitimate for an 
association to formulate standard product lines and as an integral part 
thereof for competitors to agree on what standard product lines 
should be. Just as in the case of standardization, it is illegal if the 
program is misused in order to fix or raise prices, restrict production, 
or eliminate competition, and it is obvious that an agreement by the 
participants in such a plan to adhere to the standard product lines and 
to limit their manufacture thereto is an illegal restraint of trade. 



REFERENCES 



268 U. a 663 (1925). 

2 S. D. Ohio, 1923. 

3 Docket 2199. 

4 Docket 4551. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic or microfilm copies of articles in magazines that are available may be 
obtained from The Library of Congress, Washington, D. C., or from the New York 
Public Library, New York, N. Y., at prevailing rates. 



Acoustical Society of America, Journal 

17, 4 (Apr., 1946) 
Stereophon Sound Recording System (p. 356) 

Bell Laboratories Record 

24, 4 (Apr., 1946) 
A Wide-Angle Fastax (p. 139) 
Demountable Soundproof Rooms (p. 150) 

Electronics 

19, 5 (May, 1946) 
Where Color Television Stands (p. 104) 

International Photographer 

18, 3 (Apr., 1946) 
Two-Color Photography (p. 5) 
Film Patrol for Race Tracks (p. 12) 

International Projectionist 

21, 4 (Apr., 1946) 
50-Mm Film Tests Seen as Industry Effort to 

Neutralize Competitive Threat (p. 8) 
A New Super H.I. Positive Carbon (p. 10) 

The Barrel-Type, Cyclindrical Shutter (p. 14) 
The W.U. Concentrated-Arc Lamp (p. 16) 
Basic Radio and Television Course, Pt. 22 
Transmitter Components (p. 20) 

Motion Picture Herald 

163, 4 (Apr. 27, 1946) 
Color Television Transmitted via Coaxial Cable 

(p. 51) 
526 



H. B. LEE 



J. H. WADDELL 
W. S. GORTON 



D. G. FINK 



K. MARCUS 



M. T. JONES, R. J. ZAVESKY, 

AND W. W. LOZIER 
L. DANIELS 



M. BERINSKY 



SOCIETY ANNOUNCEMENTS 527 

Philips Technical Review 

. 8, 2 (Feb., 1946) 

The Formation of Stereophonic Images (p. 51) K. DE BOER 

RCA Review 

7, 1 (Mar., 1946) 

Improved Cathode-Ray Tubes with Metal- D. W. EPSTEIN AND L. 

Backed Luminescent Screens (p. 5) PENSAK 
Local Oscillator Radiation and Its Effect on 

Television Picture Contrast (p. 32) E. W. HEROLD 

Image Orthicon Camera ( p . 67) R . D . KELL AN D G . C . SZIKLAI 



SOCIETY ANNOUNCEMENTS 



RE-ESTABLISHMENT OF MIDWEST SECTION 

The Board of Governors of the Society, at its meeting on May 5, 1946, approved 
the petition from qualified members in the Chicago area to re-establish the Mid- 
west Section. Activities of the Section had been suspended since January, 1943 
owing to the absence of many members serving with the Armed Forces. 

Early this year, a group of Society members in the area held an organization 
meeting to discuss the possibilities of re-establishing the Section, elected tempo- 
rary officers, and took steps to re-organize the Section. The present officers are: 

A. SHAPIRO, Chairman 

C. E. PHILLIMORE, Vice-Chairman 
t R. E. LEWIS, Secretary-Treasurer 

G. W. COLBURN, Manager S. A. LUKES, Manager 

E. W. D'ARCY, Manager R. T. VANNIMAN, Manager 

O. B. DEPUE, Manager C. H. STONE, Manager 

Several technical meetings have already been held and the enthusiasm of both 
officers and members promises a successful organization and future for the Section. 
Members in the Chicago area who are interested in receiving notices of meetings 
and have not received previous notices should communicate with the Secretary- 
Treasurer, R. E. Lewis, Armour Research Foundation, Illinois Institute of 
Technology, Technology Center, Chicago 16, 111. 

Previous speakers at Section meetings have been F. E. Carlson, General Electric 
Co., Cleveland, Ohio, who recently discussed "Tungsten Filament Light Sources 
for Picture Projection and Sound Reproduction," Hugh Knowles, Jensen Radio 
Manufacturing Co., Chicago, who discussed "Loudspeaker Design Trends," W. 
T. Strickland and E. E. Bickel, both of the Simpson Optical Manufacturing Co., 
Chicago, who discussed ' Coated Optics" and "Correlation of Optical and Me- 
chanical Design," and H. C. Froula, Armour Research Foundation, discussed 
"Electron Microscope Photography." Many interesting meetings are planned 
by the Board of Managers for the coming Fall season. 

The Board of Governors extends a hearty welcome to the new Section, and 
assures it of the utmost cooperation. 



528 FIFTY-NINTH TECHNICAL CONFERENCE Vol 46, No. 6 

MAY TECHNICAL CONFERENCE 

One of the most successful meetings ever held by the Society was concluded 
on May 10, 1946, when the 5-day Fifty- Ninth Semi- Annual Technical Con- 
ference was adjourned by President D. E. Hyndman, at the Hotel Pennsylvania, 
New York. Over 400 paid registrations were recorded. All of the pre-war social 
functions were resumed and contributed in large measure to the success of the 
Conference. Both the Get-Together Luncheon, at which the guest speaker 
was William F. Rodgers, Vice-President in charge of Distribution, Loew's In- 
corporated, New York, and the Dinner-Dance were outstanding. At the latter, 
a citation and scroll were presented to Thomas Armat in recognition of the fiftieth 
anniversary of the first exhibition of motion pictures in a theater. They were re- 
ceived by Lt. Brooke Armat. 

A citation and scroll were also presented to the Warner Brothers in recognition 
of then- pioneering courage and efforts in the development of sound recording and 
sound reproduction for motion pictures. Major Albert Warner received the scroll 
in behalf of himself and his brothers. 

Because of the difficulties in obtaining sufficient papers by the Papers Com- 
mittee in advance of the Conference to permit mailing of a final program to the 
membership, the editors are publishing in this issue of the JOURNAL the complete 
program as followed for the 5-day meeting. All technical sessions were held in 
the Salle Moderne of the Hotel Pennsylvania, unless otherwise indicated. The 
papers listed will be scheduled for early appearance in the JOURNAL. 



Monday, May 6, 1946 

Open Morning. 

12:30 p.m. Penn Top: Get-Together Luncheon. Guest speaker, William F. 
Rodgers, Vice-President in charge of Distribution, Loew's In- 
corporated, New York. 

2:00 p.m. Opening business and technical session of Conference. 
Frank E. Cahill, Jr., Chairman 

Session opened with a 35-mm motion picture short. 

Welcome by President Donald E. Hyndman. 

Report of the Convention Vice-President, W. C. Kunzmann. 

"Sensitometric Control of the Duping Process," by J. P. Weiss, 

Photo Products Dept., E. I. duPont de Nemours and Co., Parlin, 

N.J. 
"Rapid Test for Ferricyanide Bleach Exhaustion," by L. E. Varden 

and E. G. Seary, Pavelle Color, Inc., New York. 
"A Processing Control Sensitometer, " by G. A. Johnson, Eastman 

Kodak Co., Rochester, N. Y. 



June, 1946 FlFTY-NlNTH TECHNICAL CONFERENCE 529 

"An Improved Method for the Determination of Hydroquinone and 
Metol in Photographic Developers," by H. L. Baumbach, Para- 
mount Pictures, Inc., Hollywood. 

"Application of Methyl Ethyl Ketone to the Analysis of Developers 
for Elon and Hydroquinone," by V. C. Shaner and M. R. Sparks, 
Eastman Kodak Co., Hollywood. 

8 : 00 p.m. Evening Session. 

Earl I. Sponable, Chairman 

Session opened with a 16-mm motion picture short. 

"Lighting a Subject for Color Photography," by R. M. Evans 
Eastman Kodak Company, Rochester, N. Y. This was an 
illustrated lecture in joint session with the Inter-Society Color 
Council. 



Tuesday, May 7, 1946 
9:30 a.m. Morning Session. 

John L. Forrest, Chairman 

Session opened with a 35-mm motion picture short. 

"Color for Producer and Consumer," by I. H. Godlove, General 

Aniline and Film Corp., Easton, Pa. 
"Light Sources and Colored Objects," by R. M. Evans, Eastman 

Kodak Co., Rochester, N. Y. 

These two papers and the lecture above were arranged through 

the courtesy of the Inter-Society Color Council. 

2:00 p.m. RCA-NBC television demonstration at Radio City, New York. 
George L. Beers, Chairman, and in charge of arrangements. 

8:00 p.m. Evening Session. 

Nathan D. Golden, Chairman 

Session opened with a 35-mm motion picture short. 

"The Agfacolor Negative-Positive Method for Professional Motion 
Pictures," presented by H. C. Harsh, on behalf of the U. S. De- 
partment of Commerce. Members of the Technical Mission 
who were present are N. D. Golden, Chief, Motion Picture Sec- 
tion, U. S. Department of Commerce, sponsor of this program, 
H. W. Sachs, Remington-Rand Corp., and H. C. Harsh, Ansco. 

Wednesday, May 8, 1946 

9 : 30 a.m. A visit to DuMont's John Wanamaker Television Studios including 
a look "back stage" into studio operation. D. R. White, Chair- 
man, and in charge of arrangements. 



530 FIFTY-NINTH TECHNICAL CONFERENCE Vol 46, No. 6 

2 : 00 p.m. Afternoon Session. 

Hollis "W. Moyse, Chairman 

Session opened with a 16-mm motion picture short. 

"Report of the Committee on Motion Picture Instruction," by 
J. G. Frayne, Chairman, Electrical Research Products Div., 
Western Electric Co., Inc., Hollywood. 

"Zoom Lens for Motion Picture Cameras with Single-Barrel Linear 
Movement," by F. G. Back, Research and Development Labora- 
tory, New York. 

"A Combination 16-Mm and 35-Mm Film Processing Machine for 
Black-and- White or Color," by W. L. Prager, Solar Aircraft Co., 
Fonda Film Processing Equipment Div., Hollywood. 

"Light Control by Polarization," by J. A. Norling, Loucks and 
Norling Studios, New York. 

"Naval-Type Training Epidiascope for Universal Projection of 
Solid Objects," by Jacques Bolsey, Bol, Ltd., New York. 

7:15 p.m. Georgian Room ( Foyer} : A social hour with the Board of Governors 
preceding the Dinner-Dance. 

8:30 p.m. Georgian Room: Fifty-Ninth Semi- Annual Technical Conference 
Dinner-Dance. 

President Donald E. Hyndman, presiding 

Citation and scroll presented to Warner Brothers, represented by 

Major Albert Warner. 
Citation and scroll presented to Thomas Armat, represented by 

Lieutenant Brooke Armat. 
Dancing and entertainment until 1 : 30 a.m. 

Thursday, May 9, 1946 

Open morning. 
2 : 00 p.m. Afternoon Session. 

John G. Frayne, Chairman 

Session opened with a 35-mm motion picture short. 

"Modernization Desires of a Major Studio," by L. L. Ryder, Para- 
mount Pictures, Inc., Hollywood. 

"A New Professional 16-Mm Camera and Sound Recorder," by 
J. A. Maurer, J. A. Maurer, Inc., Long Island City, N. Y. 

"Factors Governing the Frequency Response of a Variable- Area 
Film Recording Channel," by M. Rettinger and K. Singer, RCA 
Victor Div., Radio Corporation of America, Hollywood. 

"Post- War Test Equipment for Theater Servicing," by Edward 
Stanko and P. V. Smith, RCA Service Co., Camden, N. J. 

"Dubbing and Post-Synchronization Studios," by W. A. Mueller, 
Warner Bros. Pictures, Inc., Burbank, Calif. 



June, 1946 FiFTY-NlNTH TECHNICAL CONFERENCE 531 

8 : 00 p.m. Evening Session. 

Lawrence W. Davee, Chairman 

Session opened with a 35-mm motion picture short. 

"An Improved 200-Mil Push-Pull Variable-Density Modulator," 
by J. G. Frayne and T. B. Cunningham, Electrical Research 
Products Div., Western Electric Co., Hollywood. 

"Tone Control for Rerecording," by C. O. Slyfield, Walt Disney 
Productions., Burbank, Calif. 

"The High Cost of Poor Projection," by C. E. Lewis, Showman's 
Trade Review, New York. 

"Characteristics and Applications of Concentrated Arc Lamps," by 
W. D. Buckingham and C. R. Deibert, The Western Union 
Telegraph Co., Electronics Div., Water Mill, New York. 

Friday, May 10, 1946 
9:30 a.m. Morning Session. 

Frank E. Carlson, Chairman 

Session opened with a 35-mm motion picture short. 

"Technical Problems of Film Production for the Navy's Special 
Training Devices," by H. S. Monroe, Atlas Educational Film Co., 
Oak Park, 111. 

"The Photometric Calibration of Lens Apertures," by A. E. Mur- 
ray, Bausch and Lomb Optical Co., Rochester, N. Y. 

Report of Engineering Vice-President, J. A. Maurer, J. A. Maurer, 
Inc., Long Island City, N. Y. 

"Report of Committee on Standards," by F. T. Bowditch, Chair- 
man, National Carbon Co., Cleveland, Ohio. 

"Report of Committee on 16-Mm and 8-Mm Motion Pictures," by 
D. F. Lyman, Chairman, Eastman Kodak Company, Rochester, 
N. Y. 

"Report of the Subcommittee on 16-Mm Splices," by W. H. 
Offenhauser, Jr., Chairman, Columbia Broadcasting System, 
Inc., New York. 

"Report of Committee on Studio Lighting," by C. W. Handley, 
Chairman, National Carbon Co., Los Angeles, California. 

2:00 p.m. Afternoon Session. 

Ralph B. Austrian, Chairman 

Session opened with a 35-mm motion picture short. 

"Report of Committee on Television Projection Practice," by P. J. 

Larsen, Chairman, Washington, D. C. 
"A Unified Approach to the Performance of Photographic Film, 

Television Pickup Tubes, and the Human Eye," by Albert Rose, 

RCA Laboratories, Princeton, N. J. 



532 FIFTY-NINTH TECHNICAL CONFERENCE Vol 46, No. 6 

"Color Television," by P. C. Goldmark, Columbia Broadcasting 

System, Inc., New York. 
"A New Film for Photogr